Fingerprint encryption and verification method and device based on visual password and LSB watermark
Through the combination of visual password and LSB watermarking technology, a shared image embedded in employee images is generated, which solves the problems of data insecure and insufficient decryption computing capabilities in fingerprint authentication, and achieves fingerprint verification with high security and low computing burden.
Patent Information
- Application Number
- CN202411295681.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-18
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-09-18
AI Technical Summary
The existing fingerprint authentication schemes have problems with the security risks of fingerprint feature information and insufficient decryption computing capabilities, especially the inability to handle complex decryption algorithms.
Using visual password (VCS) and LSB watermarking technology, two shared images are generated by pre-processing and encrypting employee fingerprint images, one of which is embedded in employee images and the other is stored on the enterprise side to ensure that even if the database data is leaked It is also impossible to decode the correct fingerprint and reduce the burden of decryption calculation.
Improves the security of fingerprint data, resists decryption attacks, and effectively performs the decryption process on lightweight fingerprint scanners, reducing the computing burden.
Smart Images

Figure CN118823884B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of identity authentication, and more particularly to a fingerprint encryption and verification method and device based on visual password and LSB watermark. Background Art
[0002] Confidential areas of an enterprise or organization require fingerprint authentication for access. One existing approach is to store the fingerprints of all authorized personnel in a corporate database and use fingerprint scanners at entrances. This solution requires individuals to first scan their fingerprint, and access is granted only if the scanned fingerprint matches one in the database.
[0003] However, this approach presents two challenges. First, there's the security risk of fingerprint feature information being stolen. This means that fingerprints stored in the database could be accessed by attackers. Even if fingerprints are encrypted before storage, attackers with sufficient time and computing power can still decode the correct fingerprint, or at least a portion of it. Second, there's the computational power required for decryption. Decryption is typically performed on a lightweight, compact fingerprint scanner. However, such devices may not be able to handle complex decryption algorithms due to the significant computing power required.
[0004] Therefore, this application provides a fingerprint encryption and verification method and device based on visual cryptography and LSB watermark to solve the above problems. Summary of the Invention
[0005] The purpose of this application is to provide a fingerprint encryption and verification method and device based on visual cryptography and LSB watermarking, so as to solve the problems of data insecurity and complex encryption and decryption operations in existing fingerprint identity authentication schemes. By combining visual cryptography and LSB watermarking technology, the security of fingerprint data can be improved while reducing the encryption and decryption operation burden.
[0006] In a first aspect, the present application provides a fingerprint encryption method based on visual password and LSB watermark, including: obtaining an employee image, employee information and an employee fingerprint image; preprocessing the employee fingerprint image; performing VCS encryption on the preprocessed employee fingerprint image to generate two shared images: a first shared image and a second shared image; embedding the first shared image and the employee information into the employee image as an authentication image through LSB watermark technology, storing the authentication image on the employee side, and storing the second shared image on the enterprise side.
[0007] Using the above technical solution, the employee's fingerprint image is preprocessed when the fingerprint is recorded. The preprocessed employee fingerprint image is then encrypted through VCS to generate two shared images. The two shared images can be superimposed to display a fingerprint image that can be identified by the naked eye. One of the shared images uses LSB watermark technology to secretly embed the least significant bit of the employee image and is stored on the employee's end device (such as the employee's smartphone). The other shared image is stored in the enterprise-end database for fingerprint verification. The combination of VCS and LSB watermark technology ensures that even if the data stored in the database is leaked, attackers cannot decode the correct fingerprint, while ensuring that the computing power required for decryption is relatively low.
[0008] In a possible implementation, preprocessing the employee fingerprint image includes: randomly performing normalization, directional filtering enhancement, binarization, and size adjustment on the employee fingerprint image.
[0009] In a possible implementation, performing VCS encryption on the pre-processed employee fingerprint image to generate two shared images includes: converting each pixel in the pre-processed employee fingerprint image into two sub-pixels, and storing the two sub-pixels in the two shared images respectively.
[0010] In a possible implementation, each pixel in the preprocessed employee fingerprint image is converted into two sub-pixels, including: for each pixel, randomly selecting a first case and a second case; when the first case is selected, the white pixel is converted into sub-pixel 10 and sub-pixel 10, and the black pixel is converted into sub-pixel 10 and sub-pixel 01; when the second case is selected, the white pixel is converted into sub-pixel 01 and sub-pixel 01, and the black pixel is converted into sub-pixel 01 and sub-pixel 10.
[0011] In one possible implementation, the first shared image is combined with the employee information and embedded into the employee image as an authentication image through LSB watermark technology, including: using the employee information as a seed of a random number generator to generate a random sequence, selecting whether to flip the first shared image according to the random sequence, and when flipping is selected, selecting a flipping row according to the random sequence, and horizontally flipping the flipping row in the first shared image; randomly selecting an embedding layer of the employee image based on the employee information, the embedding layer including: a red layer, a green layer, and a blue layer; and embedding the first shared image into the embedding layer of the employee image to obtain an authentication image.
[0012] In a second aspect, the present application provides a fingerprint verification method based on visual cryptography and LSB watermarking, which corresponds to the fingerprint encryption method, including: obtaining employee information and an authentication image obtained by the above-mentioned fingerprint encryption method from the employee side; extracting a first shared image from the authentication image through the employee information; obtaining a second shared image from the enterprise side through the employee information; performing joint decryption on the first shared image and the second shared image to obtain a fingerprint verification image; obtaining the actual fingerprint of the employee, comparing the actual fingerprint with the fingerprint verification image, and calculating the pixel matching or mismatching degree. When the matching degree of the actual fingerprint is greater than a threshold value T1 or the mismatching degree is lower than a threshold value T2, the fingerprint verification is deemed to have passed.
[0013] In one possible implementation, extracting a first shared image from the authentication image using the employee information includes: calculating the embedded layer of the authentication image using the employee information, and extracting the first shared image from the embedded layer of the authentication image; calculating whether the first shared image is flipped using the employee information, and if so, calculating the flipped rows according to the employee information, and flipping and restoring the flipped rows in the first shared image.
[0014] In one possible implementation, the first shared image and the second shared image are jointly decrypted to obtain a fingerprint verification image, including: performing a bitwise OR operation on sub-pixels in the first shared image and the second shared image to obtain a merged image; inverting the pixels in the merged image to obtain a merged inverted image; extracting the larger sub-pixel from every two sub-pixels of the merged inverted image as the original pixel, and generating a fingerprint binarization image through the original pixel as the fingerprint verification image.
[0015] In a third aspect of the present application, a computer program product is provided, comprising: computer software instructions, which can be loaded by a processor to implement the fingerprint encryption method or the fingerprint verification method as described above.
[0016] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium has program instructions. When the program instructions are directly or indirectly executed, the fingerprint encryption method or the fingerprint verification method as described above is implemented.
[0017] Compared with the existing technology, the present application has the following beneficial effects: First, by combining VCS with LSB watermark technology, the security of fingerprint data is guaranteed. First, the employee fingerprint image is encrypted through VCS to form multiple shared images. The information contained in the multiple shared images is much larger than the information contained in the original fingerprint image. It follows the confidentiality of information theory and can resist attacks from future decryption technologies to a certain extent. Moreover, the black and white pixel ratios in the shared images are the same, and no biased information can be obtained, which has good security. Secondly, by combining LSB watermark technology to covertly embed one of the shared images into the employee image, the eavesdropper will not even realize that a secret is being shared. Second, by combining VCS with LSB watermark technology, the adaptability of decryption resources is guaranteed. This solution forms two shared images through VCS encryption. The real fingerprint image can be displayed by superimposing the two shared images, which greatly reduces the computational burden. Moreover, LSB watermark technology is relatively mature, and the amount of computation can also be accepted by lightweight fingerprint scanners. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:
[0019] Figure 1 The flowchart of the fingerprint encryption method based on visual password and LSB watermark is shown in FIG.
[0020] Figure 2 Directional diagram for directional filtering enhancement;
[0021] Figure 3 Employee fingerprint images before and after preprocessing;
[0022] Figure 4 Schematic diagram for VCS encryption;
[0023] Figure 5 Flowchart of fingerprint verification method based on visual password and LSB watermark;
[0024] Figure 6 is a schematic diagram of a first shared image and a second shared image;
[0025] Figure 7 Schematic diagram of fingerprint verification image generation. DETAILED DESCRIPTION
[0026] In order to make the objectives, technical solutions and advantages of this application more clear, the present application is further described in detail below in conjunction with examples and drawings. The schematic implementation methods of this application and their descriptions are only used to explain this application and are not intended to limit this application.
[0027] First, in order to facilitate understanding of the technical solutions of the embodiments of the present application, a brief introduction to the relevant technologies and terms of the present application is given as follows.
[0028] VCS (Visual Cryptography Scheme) is an encryption technology that splits a secret image into multiple encrypted shared images. These shared images may visually appear to be random black and white pixels. Only when several specific shared images are superimposed together in the correct way can the observer see the original secret image through the human visual system.
[0029] LSB (Least Significant Bit) watermarking is a digital watermarking technology that embeds watermark information by modifying the least significant bit of the host data (such as an image, audio, or video file). The effect is to achieve covert embedding of the watermark without significantly affecting the quality of the original data.
[0030] Currently, fingerprint authentication schemes face two major challenges: first, fingerprint data security, which stems from potential risks in fingerprint databases, potentially leading to unauthorized access and misuse of sensitive information; and second, a lack of decryption resources, which stems from the limited computing power of lightweight fingerprint scanners, which are unable to decipher complex fingerprint passwords. Based on these issues, this scheme proposes a fingerprint encryption and verification method and device based on visual cryptography and LSB watermarking. These aims are: first, to ensure fingerprint data security, ensuring that even if the database is compromised, the stored fingerprint cannot be effectively decoded; and second, to ensure that computing resources are adaptable, so that the decryption process can be efficiently performed on lightweight fingerprint scanners, reducing the computational burden.
[0031] See Figure 1 As shown, Figure 1 This is a flow chart of a fingerprint encryption method based on visual cryptography and LSB watermarking. The fingerprint encryption method includes: S110, obtaining employee images, employee information and employee fingerprint images; S120, preprocessing the employee fingerprint images; S130, performing VCS encryption on the preprocessed employee fingerprint images to generate two shared images: a first shared image and a second shared image; S140, embedding the first shared image and employee information into the employee image as an authentication image using LSB watermarking technology, storing the authentication image on the employee side, and storing the second shared image on the enterprise side.
[0032] Specifically, when entering fingerprints, the employee's fingerprint image is preprocessed, including normalization, directional filtering enhancement, binarization, and size adjustment. The preprocessed employee fingerprint image is then encrypted through VCS to generate two shared images. The two shared images can be superimposed to display a fingerprint image that can be recognized by the naked eye. One of the shared images uses LSB watermarking technology to secretly embed the least significant bit of the employee image and is stored on the employee's device (such as the employee's smartphone), while the other shared image is stored in the enterprise database for fingerprint verification.
[0033] This solution offers several improvements: First, it combines VCS with LSB watermarking technology to ensure fingerprint data security. First, VCS encrypts employee fingerprint images to create multiple shared images. These images contain significantly more information than the original employee fingerprint image, adhering to information-theoretic confidentiality and providing a degree of resistance to future decryption attacks. Furthermore, the black-white pixel ratios in the shared images are consistent, preventing any biased information from being derived, resulting in superior security. Second, by using LSB watermarking technology to covertly embed one of the shared images within the employee image, an eavesdropper is unaware that a secret is being shared. Second, by combining VCS with LSB watermarking, the adaptability of decryption resources is ensured. This solution uses VCS encryption to create two shared images. Overlaying these two shared images reveals the true fingerprint image, significantly reducing the computational burden. Furthermore, LSB watermarking technology is relatively mature, and the computational complexity is acceptable for lightweight fingerprint scanners.
[0034] It should be explained that the aforementioned information-theoretic secrecy means that if a cryptographic system is secure against an adversary with unlimited computing resources and time, then it has information-theoretic security (also known as unconditional security). VCS is proven to be an information-theoretic secure protocol that is resistant to future advances in computing. As shown in Equations (1) and (2), in terms of entropy, knowledge of the ciphertext has no effect on the entropy of the plaintext, that is, knowledge of the ciphertext does not help determine the plaintext.
[0035] (1)
[0036] (2)
[0037] Where H(P|C) is the plaintext entropy of a given ciphertext, H(K) is the entropy of the key, and H(P) is the entropy of the plaintext, or the length of the key is greater than or equal to the length of the message. The size of the VCS share is equal to or greater than the size of the plaintext image. For example, in the visual cryptography scheme VCS in this scheme, the shared image is twice the size of the original image. Therefore, VCS adheres to information-theoretic confidentiality and is secure and reliable.
[0038] In a possible implementation, step S120 , pre-processing the employee fingerprint image, includes: pre-processing the employee fingerprint image including normalization, directional filtering enhancement, binarization, and size adjustment.
[0039] Specifically, the fingerprint region is cropped before preprocessing: an edge detection algorithm is applied to accurately locate the edge contours of the fingerprint. Morphological operations are then used to fill the detected edges to form a complete fingerprint region. The original image is then cropped based on the filled fingerprint region to obtain an image containing only the fingerprint region. The image containing only the fingerprint region is preprocessed as follows.
[0040] 1. Normalization: In order to balance the grayscale value of the employee fingerprint image to between 0 and 1, the grayscale value of the image needs to be normalized. G , the image resolution is .
[0041] First calculate the mean of the image G M and variance G V , as shown in formula (3):
[0042] (3)
[0043] in, 、 are the number of rows and columns of the image, For the image Rank Grayscale value of the pixel in the column;
[0044] The minimum grayscale value of the employee fingerprint image is recorded as , the maximum gray value is recorded as , as calculated by formula (4) and , and then adjust the overall grayscale value distribution of the employee fingerprint image:
[0045] (4)
[0046] As shown in formula (5), the mean value of the image is finally used G M 、 Variance of the image G V Normalize the image to get the normalized image :
[0047] (5).
[0048] 2. Directional filtering enhancement: 8 directional filters are used to enhance employee fingerprint images, such as Figure 2 Shown is a directional diagram of directional filtering enhancement.
[0049] 3. Binarization: Using the local adaptive threshold binarization method, this method can adaptively set the threshold for different areas of the image, effectively remove the residual noise in the image, so that the fingerprint texture information can be more clearly displayed.
[0050] 4. Size adjustment: Standardize the size of the binary employee fingerprint image and set the resolution to .
[0051] It should be noted that the employee fingerprint images before and after preprocessing are as follows Figure 3 As shown, Figure 3 The left side shows the employee’s fingerprint image before preprocessing. Figure 3 The right side shows the preprocessed employee fingerprint image. This preprocessing improves the quality of the collected employee fingerprint image, thereby enhancing the accuracy and success rate of fingerprint verification.
[0052] In a possible implementation, S130 performs VCS encryption on the preprocessed employee fingerprint image to generate two shared images, including: converting each pixel in the preprocessed employee fingerprint image into two sub-pixels, and storing the two sub-pixels in the two shared images respectively.
[0053] Specifically, this solution performs VCS encryption on the pre-processed employee fingerprint image. VCS is performed on the binary employee fingerprint image to generate two shared images. One copy is stored in the enterprise database, and the other is embedded in the employee image using LSB watermarking technology.
[0054] The following describes the process of VCS encryption to form multiple shared images. For a given original image, it is encrypted as shown in formula (6). T Divided into n shares, and requires k ( k ≤ n ) shares ( S 1, S 2… S k ) reconstructs the image. Each share is randomly generated, and the original image T can be viewed as some function of these shares.
[0055] (6)
[0056] Each pixel in the original image is encrypted into two sub-pixels, which are stored in two shared images. The distribution of white and black pixels is determined randomly. Because different pixels in the original image are encrypted using independent random selections, neither shared image provides any information about the original pixel.
[0057] The value of the original pixel can be determined by superimposing the two shared images. If the original pixel is black, two black sub-pixels are obtained; if the original pixel is white, one black sub-pixel and one white sub-pixel are obtained. Therefore, the width of the reconstructed image will be twice that of the original image, and the contrast will be lost by 50%. This can be avoided by selecting one of every two sub-pixels. In a binary image, 0 represents black and 1 represents white. The black sub-pixel is
[00] and the white sub-pixel is
[01] ; by taking the maximum value of two adjacent sub-pixels, the reconstructed image can be reduced to the size of the original image while also improving the contrast.
[0058] In this solution, a (2,2) VCS is chosen. The (2,2) here represents the threshold characteristic of the solution, which means that at least two of the two shared images are required to decrypt the hidden information. Because this solution's usage scenario requires one shared image on the employee side and one shared image on the enterprise side, two shared images are required, and both must be used to decrypt the fingerprint image. While VCS technology can generate more shared images, this solution is limited to two objects and considers the limited memory of many IoT devices used for decryption. Therefore, this solution uses a (2,2) VCS solution, which requires less memory.
[0059] In one possible implementation, Figure 4 As shown, Figure 4 Schematic diagram of VCS encryption. Each pixel in the preprocessed employee fingerprint image is converted into two sub-pixels, including: for each pixel, randomly selecting a first case and a second case; when the first case is selected, white pixels are converted into sub-pixel 10 and sub-pixel 10, and black pixels are converted into sub-pixel 10 and sub-pixel 01; when the second case is selected, white pixels are converted into sub-pixel 01 and sub-pixel 01, and black pixels are converted into sub-pixel 01 and sub-pixel 10.
[0060] Specifically, in a binary image, 0 is black and 1 is white. The employee fingerprint binary image of pixels is generated with a length of A random 0 / 1 sequence matrix is created, where each element in the sequence matrix is randomly generated as 0 or 1. For each original pixel in the employee fingerprint binary image, if the element at the corresponding position in the sequence matrix is 0, it can be considered as the first case: when the original pixel is white 1, two sub-pixels s1a=
[10] and s1b=
[10] are generated; when the original pixel is black 0, two sub-pixels s0a=
[10] and s0b=
[01] are generated. If the element at the corresponding position in the sequence matrix is 1, it can be considered as the second case: when the original pixel is white 1, two sub-pixels s1a=
[01] and s1b=
[01] are generated; when the original pixel is black 0, two sub-pixels s0a=
[01] and s0b=
[10] are generated. s1a / s0a are stored in the first shared image, and s1b / s0b are stored in the second shared image. When all original pixels in the employee fingerprint binary image are converted, two shared images are generated.
[0061] In one possible implementation, S140, embedding the first shared image and employee information into the employee image as an authentication image through LSB watermark technology, including: using the employee information as a seed of a random number generator to generate a random sequence, selecting whether to flip the first shared image according to the random sequence, and when flipping is selected, selecting a flipping row according to the random sequence, and horizontally flipping the flipping row in the first shared image; randomly selecting an embedding layer of the employee image through the employee information, the embedding layer including: a red layer, a green layer, and a blue layer; embedding the first shared image into the embedding layer of the employee image to obtain an authentication image.
[0062] Specifically, the employee's shared image must be transmitted from the employee's device to the verification device. Although the shared image is encrypted, broadcasting a secret is unwise. Therefore, this scheme uses LSB watermarking to hide the shared image within the employee's image, obscuring the shared information from any eavesdropper. Any eavesdropper only sees an innocent image being transmitted and is unaware that it contains important shared information.
[0063] In this solution, a combination of a standard LSB watermark and a variant watermark is used for embedding. The variant watermark flips the shared image from left to right before embedding it into the employee image. This combination of standard LSB and variant watermarks uses employee information, such as the employee ID, as a seed for a random number generator to generate a random sequence. The random sequence is then used to determine whether to perform a flip. If flipping is required, the random sequence is used as the row to be flipped.
[0064] Since employee images are RGB images, it's also necessary to determine the layer in which to embed the LSB watermark. This can be determined by performing a modulo 3 operation on the employee ID. For example, 0 represents the red layer, 1 represents the green layer, and 2 represents the blue layer. Using LSB watermarking, the shared image is embedded in the designated layer of the employee image, serving as the authentication image. This authentication image, along with other employee information such as name, ID, and department, is stored on the employee's smartphone for verification purposes.
[0065] It should be noted that the employee's smartphone will function as a key card. The employee can scan their phone against an RFID / NFC scanner, which then transmits the authentication image and employee information stored on the smartphone to the scanner. This can be used to provide a lower level of authentication for access to less critical corporate data and locations. The only requirement for access to this data and locations is proof that the individual is, in fact, an employee of the company. Access to more critical corporate data and locations requires subsequent fingerprint verification in addition to smartphone authentication.
[0066] See Figure 5 As shown, Figure 5 This is a flow chart of a fingerprint verification method based on visual cryptography and LSB watermarking. The method includes: S210, obtaining employee information and an authentication image obtained using the fingerprint encryption method described above from the employee; S220, extracting a first shared image from the authentication image using the employee information; S230, obtaining a second shared image from the enterprise using the employee information; S240, jointly decrypting the first and second shared images to obtain a fingerprint verification image; S250, obtaining the employee's actual fingerprint, comparing the actual fingerprint with the fingerprint verification image, and calculating the pixel match or mismatch. When the actual fingerprint's match is greater than a threshold value T1 or the mismatch is less than a threshold value T2, the fingerprint verification is considered successful.
[0067] In one possible implementation, S220, extracting a first shared image from an authentication image using employee information, includes: calculating an embedded layer of the authentication image using employee information, and extracting the first shared image from the embedded layer of the authentication image; calculating whether the first shared image is flipped using employee information, and if so, calculating the flipped rows according to the employee information, and performing flip recovery on the flipped rows in the first shared image.
[0068] In a possible implementation, S240, a fingerprint verification image is obtained by jointly decrypting the first shared image and the second shared image, including: performing a bitwise OR operation on sub-pixels in the first shared image and the second shared image to obtain a merged image; inverting the pixels in the merged image to obtain a merged inverted image; extracting a larger sub-pixel from every two sub-pixels of the merged inverted image as an original pixel, and generating a fingerprint binarization image through the original pixels as a fingerprint verification image.
[0069] Specifically, if Figure 6 and Figure 7 As shown, Figure 6 is a schematic diagram of a first shared image and a second shared image, Figure 7 Schematic diagram of fingerprint verification image generation. After receiving the employee data (including employee ID) and authentication image stored in the employee’s smartphone, the scanner first extracts the second shared image corresponding to the employee ID from the enterprise database based on the employee ID, such as Figure 6 (b) The mod3 value of the employee ID indicates the embedding layer of the employee image. The first shared image is extracted from the embedding layer of the employee image using the extraction algorithm, as shown in Figure 6 (a). The first shared image and the second shared image are superimposed and merged, as shown in Figure 7 (a) is shown. Decrypt the superimposed and merged image to obtain the fingerprint verification image, as shown in Figure 7 (b) Finally, the employee's actual fingerprint is compared with the fingerprint verification image. Verification is performed by performing an XOR operation on the fingerprint verification image matrix and the actual fingerprint matrix to determine the number of pixel / bit mismatches or matches. If the number of matches is greater than threshold T1 (90%) or the number of mismatches is less than threshold T2 (10%), the fingerprint verification is considered successful and further identity verification can be performed.
[0070] As shown in Table 1, this application provides a time efficiency analysis of the above method.
[0071] Fingerprint image encryption storage time, as shown in Table 1:
[0072] Table 1 Fingerprint image encryption storage time
[0073]
[0074] The fingerprint decryption authentication takes time, as shown in Table 2:
[0075] Table 2 Fingerprint decryption and authentication time
[0076]
[0077] It can be understood that this solution realizes fingerprint image encryption through the combination of visual cryptography and LSB watermark technology. The visual encryption scheme VCS, as a simple and secure method, allows the secret sharing of images, so that decryption can be performed using simple superposition; the fragile digital watermark LSB embeds invisible watermark information with tampering location capability by modifying the lowest bit of the original image, thereby realizing content integrity authentication of digital products. This solution combines the advantages of VCS and LSB technologies and applies them to the intelligent fingerprint access control system to realize the privacy protection of fingerprint feature information, providing a safe and efficient solution for fingerprint authentication in confidential locations. These technologies enhance the security of stored fingerprints and make practical implementation on lightweight fingerprint scanning devices possible.
[0078] The above describes in detail the method and effects of the embodiments of the present application. In order to facilitate better implementation of the above solutions of the embodiments of the present application, correspondingly, related equipment for cooperating in implementing the above solutions is also provided below.
[0079] An embodiment of the present application further provides a computer program product, which includes computer software instructions. The computer software instructions can be loaded by a processor to implement the processes in the methods shown in the aforementioned embodiments.
[0080] The embodiment of the present application also provides a computer-readable storage medium having program instructions. When the program instructions are directly or indirectly executed, the fingerprint entry method or the fingerprint verification method as described above is implemented.
[0081] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.
Claims
1. Fingerprint encryption method based on visual password and LSB watermark, characterized in that: include: Obtain employee images, employee information, and employee fingerprint images; Preprocessing the employee's fingerprint image; Performing VCS encryption on the pre-processed employee fingerprint image to generate two shared images: a first shared image and a second shared image; The first shared image is combined with the employee information and embedded into the employee image as an authentication image through LSB watermark technology. The authentication image is stored on the employee side, and the second shared image is stored on the enterprise side.
2. The fingerprint encryption method based on visual password and LSB watermark according to claim 1 is characterized in that: The employee fingerprint image is preprocessed, including: randomly normalizing, directional filtering enhancement, binarization and size adjustment processing the employee fingerprint image.
3. The fingerprint encryption method based on visual password and LSB watermark according to claim 1 is characterized in that: The pre-processed employee fingerprint image is encrypted by VCS to generate two shared images, including: converting each pixel in the pre-processed employee fingerprint image into two sub-pixels, and storing the two sub-pixels in the two shared images respectively.
4. The fingerprint encryption method based on visual password and LSB watermark according to claim 3 is characterized in that: Converting each pixel in the preprocessed employee fingerprint image into two sub-pixels, including: For each pixel, randomly select the first case and the second case; When the first case is selected, the white pixel is converted into sub-pixel 10 and sub-pixel 10, and the black pixel is converted into sub-pixel 10 and sub-pixel 01; When the second case is selected, the white pixel is converted into the sub-pixel 01 and the sub-pixel 01, and the black pixel is converted into the sub-pixel 01 and the sub-pixel 10.
5. The fingerprint encryption method based on visual password and LSB watermark according to claim 1 is characterized in that: Embedding the first shared image and the employee information into the employee image as an authentication image by using LSB watermark technology includes: Using the employee information as a seed of a random number generator to generate a random sequence, selecting whether to flip the first shared image according to the random sequence, and when flipping is selected, selecting a flipping row according to the random sequence, and horizontally flipping the flipping row in the first shared image; Randomly select an embedding layer of an employee image according to the employee information, wherein the embedding layer includes: a red layer, a green layer and a blue layer; The first shared image is embedded into the embedding layer of the employee image to obtain a certification image.
6. A fingerprint verification method based on visual password and LSB watermark, characterized in that: include: Acquire employee information and an authentication image obtained by the fingerprint encryption method according to any one of claims 1 to 5 from the employee terminal; extracting a first shared image from the authentication image using the employee information; Acquire a second shared image from the enterprise end through the employee information; Performing joint decryption on the first shared image and the second shared image to obtain a fingerprint verification image; The actual fingerprint of the employee is obtained, the actual fingerprint is compared with the fingerprint verification image, and the pixel matching or mismatching degree is calculated. When the matching degree of the actual fingerprint is greater than the threshold T1 or the mismatching degree is lower than the threshold T2, the fingerprint verification is deemed to be passed.
7. The fingerprint verification method based on visual password and LSB watermark according to claim 6 is characterized in that: Extracting a first shared image from the authentication image using the employee information includes: Decomposing the embedded layer of the authentication image by using the employee information, and extracting the first shared image from the embedded layer of the authentication image; It is calculated by using the employee information whether the first shared image is flipped. If flipped, the flipped row is calculated according to the employee information, and the flipped row in the first shared image is flipped and restored.
8. The fingerprint verification method based on visual password and LSB watermark according to claim 7 is characterized in that: The method of jointly decrypting the first shared image and the second shared image to obtain a fingerprint verification image includes: Performing a bitwise OR operation on sub-pixels in the first shared image and the second shared image to obtain a merged image; Inverting pixels in the merged image to obtain a merged inverted image; A larger sub-pixel is extracted from every two sub-pixels of the combined inverted image as an original pixel, and a fingerprint binary image is generated through the original pixels as a fingerprint verification image.
9. A computer program product, characterized in that The computer program product comprises: computer software instructions, which can be loaded by a processor to implement the method according to any one of claims 1 to 5 or the method according to any one of claims 6 to 8.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium has program instructions, and when the program instructions are directly or indirectly executed, the method according to any one of claims 1 to 5 or the method according to any one of claims 6 to 8 is implemented.
