WO2024007095A1 - 一种门禁系统人脸数据的安全加密方法和系统 - Google Patents

一种门禁系统人脸数据的安全加密方法和系统 Download PDF

Info

Publication number
WO2024007095A1
WO2024007095A1 PCT/CN2022/103567 CN2022103567W WO2024007095A1 WO 2024007095 A1 WO2024007095 A1 WO 2024007095A1 CN 2022103567 W CN2022103567 W CN 2022103567W WO 2024007095 A1 WO2024007095 A1 WO 2024007095A1
Authority
WO
WIPO (PCT)
Prior art keywords
encrypted
encryption
image
matrix
sub
Prior art date
Application number
PCT/CN2022/103567
Other languages
English (en)
French (fr)
Inventor
王晓东
Original Assignee
嘉兴尚坤科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 嘉兴尚坤科技有限公司 filed Critical 嘉兴尚坤科技有限公司
Priority to CN202280018231.8A priority Critical patent/CN117015953B/zh
Priority to PCT/CN2022/103567 priority patent/WO2024007095A1/zh
Publication of WO2024007095A1 publication Critical patent/WO2024007095A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
    • G07C9/00—Individual registration on entry or exit
    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
    • G07C9/00—Individual registration on entry or exit
    • G07C9/30—Individual registration on entry or exit not involving the use of a pass
    • G07C9/32—Individual registration on entry or exit not involving the use of a pass in combination with an identity check
    • G07C9/37—Individual registration on entry or exit not involving the use of a pass in combination with an identity check using biometric data, e.g. fingerprints, iris scans or voice recognition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols

Definitions

  • the invention relates to the field of data security protection, and in particular to a secure encryption method and system for face data in an access control system.
  • Access control unlocking has evolved from mechanical key unlocking to electronic password locks.
  • fingerprint access control has emerged.
  • Behind the evolution of unlocking methods is on the one hand the advancement of technology, and on the other hand people are increasingly concerned about the security of door locks and the user experience of door locks. coming higher and higher demand.
  • Access control in the form of keys and passwords is more traditional, with lower security and average experience; fingerprint access control is unlocked by the user's fingerprint, and the user does not need to carry a key or remember a password.
  • the security and experience are slightly higher than traditional door locks, but the same There will be some problems, such as a high misrecognition rate, peeling of the user's fingers, water, etc., which makes unlocking unsuccessful, and unlocking cannot be done when both hands are occupied, etc.
  • Face recognition is a biometric technology for identity recognition based on people's facial feature information. It uses a camera or camera to collect images or video streams containing faces, and automatically detects and tracks faces in the images, thereby detecting A series of related technologies for face recognition, usually also called portrait recognition and facial recognition.
  • the purpose of the present invention is to provide a secure encryption method and system for face data in an access control system in order to overcome the above-mentioned defects in the prior art.
  • a secure encryption method for face data in an access control system which is characterized by including the following steps:
  • S1 Obtain the original face image data of the access control system, and determine the area to be encrypted based on the position of the face in the image;
  • S5 Perform encryption operations on the first processing matrix and the second processing matrix respectively to obtain encryption matrices corresponding to the first processing matrix and the second processing matrix respectively;
  • S6 Use the encryption matrix to encrypt the sub-image to be encrypted to obtain encrypted face image data.
  • the facial feature data includes facial feature points and contour feature points.
  • step S3 by encoding the facial feature points and contour feature points, a first processing matrix is obtained, and the size of the first processing matrix is is M ⁇ N, where M ⁇ N is the size of the sub-image to be encrypted.
  • the size of the three-dimensional matrix is M ⁇ N ⁇ L, where M ⁇ N is the size of the sub-image to be encrypted, and L is the number of bit plane layers.
  • step S5 a fully homomorphic encryption function is used to perform encryption operations on the first processing matrix and the second processing matrix, respectively, to obtain encryption matrices corresponding to the first processing matrix and the second processing matrix respectively.
  • step S6 specifically includes:
  • the encryption matrix corresponding to the first processing matrix and the encryption matrix corresponding to the second processing matrix are superimposed to generate a ciphertext matrix, and the ciphertext matrix replaces the area to be encrypted in the original face image data to obtain encrypted face image data.
  • step S2 also includes performing pre-encryption processing on the sub-image to be encrypted.
  • the pre-encryption processing specifically includes:
  • the texture is obtained by convolving and clustering the sub-images to be encrypted through linear spatial filters and K-means clustering algorithms;
  • Image acquisition module used to obtain the original face image data of the access control system, determine the area to be encrypted based on the position of the face in the image, and extract the image data of the pre-encrypted area in the face image to generate the sub-image to be encrypted;
  • Encryption module used to generate the first processing matrix and the second processing matrix based on the sub-image to be encrypted, and perform encryption processing on the sub-image to be encrypted to obtain encrypted face image data.
  • the system also includes a pre-encryption module, which is used to pre-encrypt the sub-image to be encrypted, specifically:
  • the texture image is obtained by convolving and clustering the sub-images to be encrypted using a linear spatial filter and K-means clustering algorithm;
  • the encryption module first uses a fully homomorphic encryption function to perform encryption operations on the first processing matrix and the second processing matrix respectively to obtain encryption matrices corresponding to the first processing matrix and the second processing matrix respectively, and then The encryption matrix corresponding to the first processing matrix and the encryption matrix corresponding to the second processing matrix are superimposed to generate a ciphertext matrix, and the ciphertext matrix replaces the area to be encrypted in the original face image data to obtain encrypted face image data.
  • the present invention has the following advantages:
  • the present invention obtains the original face image data of the access control system, determines the area to be encrypted according to the position of the face in the image, extracts the image data of the pre-encrypted area in the face image, generates the sub-image to be encrypted, and generates the sub-image to be encrypted according to the location of the face in the image.
  • the encrypted sub-image generates a first processing matrix and a second processing matrix, and the sub-image to be encrypted is encrypted to obtain encrypted face image data, and the first processing matrix and the second processing matrix generated from the sub-image to be encrypted's own data are used for encryption.
  • the second processing matrix of the present invention is obtained by layering the output of the three-dimensional matrix of the sub-image to be encrypted according to the bit plane, which increases the encryption security in the three dimensions of row, column and layer, and can process key areas to obtain the ideal encryption effect;
  • the present invention Before the encryption process, the present invention first convolves and clusters the sub-image to be encrypted through the linear space filter and the K-means clustering algorithm, divides the sub-image to be encrypted into blocks, and performs JL transformation encryption to perform pre-encryption. processing, further ensuring the encryption security of face image data, and solving the existing problems of low encryption security of face images and inability to effectively achieve image privacy protection.
  • Figure 1 is a schematic flow chart of the method of the present invention.
  • the present invention provides a secure encryption method for face data in an access control system, which includes the following steps:
  • S1 Obtain the original face image data of the access control system, and determine the area to be encrypted based on the position of the face in the image;
  • S5 Perform encryption operations on the first processing matrix and the second processing matrix respectively to obtain encryption matrices corresponding to the first processing matrix and the second processing matrix respectively;
  • S6 Use the encryption matrix to encrypt the sub-image to be encrypted to obtain encrypted face image data.
  • the facial feature data includes facial feature points and contour feature points.
  • step S3 by encoding the facial feature points and contour feature points, a first processing matrix is obtained.
  • the size of the first processing matrix is M ⁇ N, where M ⁇ N is the size of the sub-image to be encrypted, and the size of the three-dimensional matrix is M ⁇ N ⁇ L, where M ⁇ N is the size of the sub-image to be encrypted, and L is the number of bit plane layers.
  • step S5 a fully homomorphic encryption function is used to perform encryption operations on the first processing matrix and the second processing matrix respectively, and encryption matrices corresponding to the first processing matrix and the second processing matrix are obtained respectively.
  • Step S6 specifically includes: superimposing the encryption matrix corresponding to the first processing matrix and the encryption matrix corresponding to the second processing matrix to generate a ciphertext matrix, and replacing the area to be encrypted in the original face image data with the ciphertext matrix, Obtain encrypted face image data.
  • the second processing matrix is obtained by layering the three-dimensional matrix output of the sub-image to be encrypted according to the bit plane, which increases the encryption security in the three dimensions of rows, columns and layers, and can process key areas to obtain ideal encryption effects.
  • step S2 also includes pre-encryption processing of the sub-image to be encrypted.
  • the pre-encryption processing specifically includes:
  • the texture is obtained by convolving and clustering the sub-images to be encrypted through linear spatial filters and K-means clustering algorithms;
  • the invention also provides a secure encryption method system for facial data in an access control system, including:
  • Image acquisition module used to obtain the original face image data of the access control system, determine the area to be encrypted based on the position of the face in the image, and extract the image data of the pre-encrypted area in the face image to generate the sub-image to be encrypted;
  • Encryption module used to generate the first processing matrix and the second processing matrix based on the sub-image to be encrypted, and perform encryption processing on the sub-image to be encrypted to obtain encrypted face image data;
  • Pre-encryption module used to pre-encrypt the sub-image to be encrypted, specifically: convolve and cluster the sub-image to be encrypted through a linear spatial filter and K-means clustering algorithm to obtain the texture image; based on the texture image to be encrypted
  • the sub-image is divided into blocks to obtain multiple image blocks; the multiple image blocks are separately encrypted by JL transformation to complete the pre-encryption process.
  • the encryption module first uses a fully homomorphic encryption function to perform encryption operations on the first processing matrix and the second processing matrix, respectively, to obtain the encryption matrices corresponding to the first processing matrix and the second processing matrix, and then converts the encryption matrix corresponding to the first processing matrix into The encryption matrix corresponding to the second processing matrix is superimposed to generate a ciphertext matrix, and the ciphertext matrix replaces the area to be encrypted in the original face image data to obtain encrypted face image data.
  • the pre-encryption module first convolves and clusters the sub-images to be encrypted through the linear spatial filter and K-means clustering algorithm, divides the sub-images to be encrypted into blocks, and performs JL transformation encryption for pre-encryption processing, which further ensures the human
  • the encryption security of face image data solves the problem of low encryption security of existing face images and the inability to effectively protect image privacy.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Databases & Information Systems (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Security & Cryptography (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computing Systems (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Human Computer Interaction (AREA)
  • Multimedia (AREA)
  • Collating Specific Patterns (AREA)

Abstract

一种门禁系统人脸数据的安全加密方法，包括以下步骤:S1:获取门禁系统的原始人脸图像数据，并根据图像中人脸的位置确定待加密区域；S2: 提取人脸图像中的预加密区域的图像数据，得到待加密子图像；S3:提取待加密子图像中的人脸特征数据，生成第一处理矩阵；S4:根据比特位面对待加密子图像进行分层，输出对应的三维矩阵作为第二处理矩阵；S5:对第一处理矩阵和第二处理矩阵分别进行加密操作，分别得到第一处理矩阵和第二处理矩阵对应的加密矩阵；S6:利用加密矩阵对待加密子图像进行加密处理，得到加密人脸图像数据，安全加密方法提高门禁系统人脸图像数据的安全性。

Description

一种门禁系统人脸数据的安全加密方法和系统 技术领域
本发明涉及数据安全保护领域，尤其是涉及一种门禁系统人脸数据的安全加密方法和系统。
背景技术
门禁解锁由机械式的钥匙解锁发展到电子密码锁，当前又出现指纹门禁，解锁方式进化的背后一方面是技术的进步，另一方面是人们对门锁的安全性以及门锁的用户体验有着越来越高的需求。钥匙、密码形式的门禁较为传统，安全性较低，体验也一般；指纹门禁通过用户指纹进行解锁，用户无需携带钥匙，也不用记密码，安全性以及体验性比传统门锁稍高，但同样会面临一些问题，比如误识率高，用户手指脱皮、有水等使得解锁不成功，双手被占用时也无法解锁等等。
人脸识别是基于人的脸部特征信息进行身份识别的一种生物识别技术，用摄像机或摄像头采集含有人脸的图像或视频流，并自动在图像中检测和跟踪人脸，进而对检测到的人脸进行识别的一系列相关技术，通常也叫做人像识别、面部识别。
随着人脸识别技术的发展和门禁解锁方式正不断变化，利用人脸识别的门禁系统快速发展，其主要原理是基于人脸图像实时采集并与预存图像数据对比进行匹配识别，门禁系统中存储着大量的人脸图像数据，从数据安全角度考虑，这很容易造成用户人脸等隐私信息的泄露，在个人隐私日益受到关注的今天，若门禁系统对人脸图像数据的处理不当，极有可能为用户巨大的困扰。
发明内容
本发明的目的就是为了克服上述现有技术存在的缺陷而提供一种门禁系统人脸数据的安全加密方法和系统。
本发明的目的可以通过以下技术方案来实现：
一种门禁系统人脸数据的安全加密方法，其特征在于，包括以下步骤：
S1：获取门禁系统的原始人脸图像数据，并根据图像中人脸的位置确定待加 密区域；
S2：提取人脸图像中的预加密区域的图像数据，得到待加密子图像；
S3：提取待加密子图像中的人脸特征数据，生成第一处理矩阵；
S4：根据比特位面对待加密子图像进行分层，输出对应的三维矩阵作为第二处理矩阵；
S5：对第一处理矩阵和第二处理矩阵分别进行加密操作，分别得到第一处理矩阵和第二处理矩阵对应的加密矩阵；
S6：利用加密矩阵对待加密子图像进行加密处理，得到加密人脸图像数据。
进一步地，所述的人脸特征数据包括五官特征点和轮廓特征点，步骤S3中，通过对五官特征点和轮廓特征点进行编码，得到第一处理矩阵，所述的第一处理矩阵的尺寸为M×N，其中M×N为待加密子图像的尺寸。
进一步地，所述的三维矩阵的尺寸为M×N×L，其中M×N为待加密子图像的尺寸，L为位面层数。
进一步地，步骤S5中，采用全同态加密函数对所述的第一处理矩阵和第二处理矩阵分别进行加密操作，分别得到第一处理矩阵和第二处理矩阵对应的加密矩阵。
更进一步地，步骤S6中，具体包括：
将第一处理矩阵对应的加密矩阵和第二处理矩阵对应的加密矩阵进行叠加，生成密文矩阵，并将密文矩阵替换原始人脸图像数据中的待加密区域，得到加密人脸图像数据。
进一步地，步骤S2还包括对待加密子图像进行预加密处理。
更进一步地，所述的预加密处理具体包括：
通过线性空间滤波器和K-均值聚类算法对待加密子图像进行卷积和聚类，得到纹理；
基于所述的纹理图像对所述的待加密子图像进行分块，得到多个图像块；
分别对多个图像块进行JL变换加密，完成预加密处理。
一种用于实现如所述的门禁系统人脸数据安全加密方法的系统，包括：
图像获取模块：用于获取门禁系统的原始人脸图像数据，根据图像中人脸的位置确定待加密区域，并提取人脸图像中的预加密区域的图像数据，生成待加密子图像；
加密模块：用于根据待加密子图像生成第一处理矩阵和第二处理矩阵，并对待 加密子图像进行加密处理，得到加密人脸图像数据。
优选地，该系统还包括预加密模块，所述的预加密模块用于对待加密子图像进行预加密处理，具体为：
通过线性空间滤波器和K-均值聚类算法对待加密子图像进行卷积和聚类，得到纹理图像；
基于所述的纹理图像对所述的待加密子图像进行分块，得到多个图像块；
分别对多个图像块进行JL变换加密，完成预加密处理。
进一步地，所述的加密模块首先采用全同态加密函数对所述的第一处理矩阵和第二处理矩阵分别进行加密操作，分别得到第一处理矩阵和第二处理矩阵对应的加密矩阵，再将第一处理矩阵对应的加密矩阵和第二处理矩阵对应的加密矩阵进行叠加，生成密文矩阵，并将密文矩阵替换原始人脸图像数据中的待加密区域，得到加密人脸图像数据。
与现有技术相比，本发明具有以下优点：
1)本发明通过获取门禁系统的原始人脸图像数据，根据图像中人脸的位置确定待加密区域，并提取人脸图像中的预加密区域的图像数据，生成待加密子图像，并根据待加密子图像生成第一处理矩阵和第二处理矩阵，并对待加密子图像进行加密处理，得到加密人脸图像数据，利用待加密子图像自身数据生成的第一处理矩阵和第二处理矩阵进行加密，对人脸图像数据执行加密隐藏避免用户隐私泄漏，实现更为安全更为缜密的信息加密方案，具有针对性强、实时加密、且安全性高不易被破解的优点；
2)本发明的第二处理矩阵通过根据比特位面对待加密子图像进行分层输出的三维矩阵得到，增加行、列和层三个维度的加密安全性，能够针对重点区域进行处理以获得理想的加密效果；
3)本发明在加密处理之前，先通过线性空间滤波器和K-均值聚类算法对待加密子图像进行卷积和聚类，对待加密子图像进行分块，并进行JL变换加密，进行预加密处理，进一步保证了人脸图像数据的加密安全性，解决了现有人脸图像加密安全性不高和无法有效实现图像隐私保护的问题。
附图说明
图1为本发明方法的流程示意图。
具体实施方式
下面结合附图和具体实施例对本发明进行详细说明。显然，所描述的实施例是本发明的一部分实施例，而不是全部实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例，都应属于本发明保护的范围。
如图1所示，本发明提供一种门禁系统人脸数据的安全加密方法，包括以下步骤：
S1：获取门禁系统的原始人脸图像数据，并根据图像中人脸的位置确定待加密区域；
S2：提取人脸图像中的预加密区域的图像数据，得到待加密子图像；
S3：提取待加密子图像中的人脸特征数据，生成第一处理矩阵；
S4：根据比特位面对待加密子图像进行分层，输出对应的三维矩阵作为第二处理矩阵；
S5：对第一处理矩阵和第二处理矩阵分别进行加密操作，分别得到第一处理矩阵和第二处理矩阵对应的加密矩阵；
S6：利用加密矩阵对待加密子图像进行加密处理，得到加密人脸图像数据。
人脸特征数据包括五官特征点和轮廓特征点，步骤S3中，通过对五官特征点和轮廓特征点进行编码，得到第一处理矩阵，第一处理矩阵的尺寸为M×N，其中M×N为待加密子图像的尺寸，三维矩阵的尺寸为M×N×L，其中M×N为待加密子图像的尺寸，L为位面层数。
具体地，步骤S5中，采用全同态加密函数对第一处理矩阵和第二处理矩阵分别进行加密操作，分别得到第一处理矩阵和第二处理矩阵对应的加密矩阵。步骤S6中，具体包括：将第一处理矩阵对应的加密矩阵和第二处理矩阵对应的加密矩阵进行叠加，生成密文矩阵，并将密文矩阵替换原始人脸图像数据中的待加密区域，得到加密人脸图像数据。第二处理矩阵通过根据比特位面对待加密子图像进行分层输出的三维矩阵得到，增加行、列和层三个维度的加密安全性，能够针对重点区域进行处理以获得理想的加密效果。
本实施例中，步骤S2还包括对待加密子图像进行预加密处理，具体地，预加密处理具体包括：
通过线性空间滤波器和K-均值聚类算法对待加密子图像进行卷积和聚类，得到纹理；
基于纹理图像对待加密子图像进行分块，得到多个图像块；
分别对多个图像块进行JL变换加密，完成预加密处理。
本发明还提供一种门禁系统人脸数据的安全加密方法系统，包括：
图像获取模块：用于获取门禁系统的原始人脸图像数据，根据图像中人脸的位置确定待加密区域，并提取人脸图像中的预加密区域的图像数据，生成待加密子图像；
加密模块：用于根据待加密子图像生成第一处理矩阵和第二处理矩阵，并对待加密子图像进行加密处理，得到加密人脸图像数据；
预加密模块：用于对待加密子图像进行预加密处理，具体为：通过线性空间滤波器和K-均值聚类算法对待加密子图像进行卷积和聚类，得到纹理图像；基于纹理图像对待加密子图像进行分块，得到多个图像块；分别对多个图像块进行JL变换加密，完成预加密处理。
加密模块首先采用全同态加密函数对第一处理矩阵和第二处理矩阵分别进行加密操作，分别得到第一处理矩阵和第二处理矩阵对应的加密矩阵，再将第一处理矩阵对应的加密矩阵和第二处理矩阵对应的加密矩阵进行叠加，生成密文矩阵，并将密文矩阵替换原始人脸图像数据中的待加密区域，得到加密人脸图像数据。
预加密模块先通过线性空间滤波器和K-均值聚类算法对待加密子图像进行卷积和聚类，对待加密子图像进行分块，并进行JL变换加密，进行预加密处理，进一步保证了人脸图像数据的加密安全性，解决了现有人脸图像加密安全性不高和无法有效实现图像隐私保护的问题。
以上所述，仅为本发明的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的工作人员在本发明揭露的技术范围内，可轻易想到各种等效的修改或替换，这些修改或替换都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应以权利要求的保护范围为准。

Claims (10)

  1. 一种门禁系统人脸数据的安全加密方法，其特征在于，包括以下步骤：
    S1：获取门禁系统的原始人脸图像数据，并根据图像中人脸的位置确定待加密区域；
    S2：提取人脸图像中的预加密区域的图像数据，得到待加密子图像；
    S3：提取待加密子图像中的人脸特征数据，生成第一处理矩阵；
    S4：根据比特位面对待加密子图像进行分层，输出对应的三维矩阵作为第二处理矩阵；
    S5：对第一处理矩阵和第二处理矩阵分别进行加密操作，分别得到第一处理矩阵和第二处理矩阵对应的加密矩阵；
    S6：利用加密矩阵对待加密子图像进行加密处理，得到加密人脸图像数据。
  2. 根据权利要求1所述的一种门禁系统人脸数据的安全加密方法，其特征在于，所述的人脸特征数据包括五官特征点和轮廓特征点，步骤S3中，通过对五官特征点和轮廓特征点进行编码，得到第一处理矩阵，所述的第一处理矩阵的尺寸为M×N，其中M×N为待加密子图像的尺寸。
  3. 根据权利要求1所述的一种门禁系统人脸数据的安全加密方法，其特征在于，所述的三维矩阵的尺寸为M×N×L，其中M×N为待加密子图像的尺寸，L为位面层数。
  4. 根据权利要求1所述的一种门禁系统人脸数据的安全加密方法，其特征在于，步骤S5中，采用全同态加密函数对所述的第一处理矩阵和第二处理矩阵分别进行加密操作，分别得到第一处理矩阵和第二处理矩阵对应的加密矩阵。
  5. 根据权利要求4所述的一种门禁系统人脸数据的安全加密方法，其特征在于，步骤S6中，具体包括：
    将第一处理矩阵对应的加密矩阵和第二处理矩阵对应的加密矩阵进行叠加，生成密文矩阵，并将密文矩阵替换原始人脸图像数据中的待加密区域，得到加密人脸图像数据。
  6. 根据权利要求1所述的一种门禁系统人脸数据的安全加密方法，其特征在于，步骤S2还包括对待加密子图像进行预加密处理。
  7. 根据权利要求6所述的一种门禁系统人脸数据的安全加密方法，其特征在于，所述的预加密处理具体包括：
    通过线性空间滤波器和K-均值聚类算法对待加密子图像进行卷积和聚类，得到纹理；
    基于所述的纹理图像对所述的待加密子图像进行分块，得到多个图像块；
    分别对多个图像块进行JL变换加密，完成预加密处理。
  8. 一种用于实现如权利要求1-7任一项所述的门禁系统人脸数据安全加密方法的系统，其特征在于，包括：
    图像获取模块：用于获取门禁系统的原始人脸图像数据，根据图像中人脸的位置确定待加密区域，并提取人脸图像中的预加密区域的图像数据，生成待加密子图像；
    加密模块：用于根据待加密子图像生成第一处理矩阵和第二处理矩阵，并对待加密子图像进行加密处理，得到加密人脸图像数据。
  9. 根据权利要求8所述的一种门禁系统人脸数据的安全加密系统，其特征在于，该系统还包括预加密模块，所述的预加密模块用于对待加密子图像进行预加密处理，具体为：
    通过线性空间滤波器和K-均值聚类算法对待加密子图像进行卷积和聚类，得到纹理图像；
    基于所述的纹理图像对所述的待加密子图像进行分块，得到多个图像块；
    分别对多个图像块进行JL变换加密，完成预加密处理。
  10. 根据权利要求8所述的一种门禁系统人脸数据的安全加密系统，其特征在于，所述的加密模块首先采用全同态加密函数对所述的第一处理矩阵和第二处理矩阵分别进行加密操作，分别得到第一处理矩阵和第二处理矩阵对应的加密矩阵，再将第一处理矩阵对应的加密矩阵和第二处理矩阵对应的加密矩阵进行叠加，生成密文矩阵，并将密文矩阵替换原始人脸图像数据中的待加密区域，得到加密人脸图像数据。
PCT/CN2022/103567 2022-07-04 2022-07-04 一种门禁系统人脸数据的安全加密方法和系统 WO2024007095A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202280018231.8A CN117015953B (zh) 2022-07-04 2022-07-04 一种门禁系统人脸数据的安全加密方法和系统
PCT/CN2022/103567 WO2024007095A1 (zh) 2022-07-04 2022-07-04 一种门禁系统人脸数据的安全加密方法和系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2022/103567 WO2024007095A1 (zh) 2022-07-04 2022-07-04 一种门禁系统人脸数据的安全加密方法和系统

Publications (1)

Publication Number Publication Date
WO2024007095A1 true WO2024007095A1 (zh) 2024-01-11

Family

ID=88574796

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/103567 WO2024007095A1 (zh) 2022-07-04 2022-07-04 一种门禁系统人脸数据的安全加密方法和系统

Country Status (2)

Country Link
CN (1) CN117015953B (zh)
WO (1) WO2024007095A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117998025A (zh) * 2024-04-07 2024-05-07 山东青橙数字科技有限公司 一种基于位阶计量运算的关键信息识别图像加密方法

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5533127A (en) * 1994-03-18 1996-07-02 Canon Information Systems, Inc. Encryption system
CN108776790A (zh) * 2018-06-06 2018-11-09 海南大学 云环境下基于神经网络的加密人脸识别方法
CN109657483A (zh) * 2018-11-27 2019-04-19 哈尔滨工业大学（深圳） 一种图像加密方法和系统
CN110633650A (zh) * 2019-08-22 2019-12-31 首都师范大学 基于隐私保护的卷积神经网络人脸识别方法及装置
CN111967033A (zh) * 2020-08-28 2020-11-20 深圳康佳电子科技有限公司 基于人脸识别的图片加密方法、装置、终端及存储介质
CN112949576A (zh) * 2021-03-29 2021-06-11 北京京东方技术开发有限公司 姿态估计方法、装置、设备及存储介质
CN113723334A (zh) * 2021-09-07 2021-11-30 合肥远康信息技术有限公司 一种金融联网综合安防系统
CN113807258A (zh) * 2021-09-17 2021-12-17 南京工业大学 一种基于神经网络与dct变换的加密人脸识别方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9852492B2 (en) * 2015-09-18 2017-12-26 Yahoo Holdings, Inc. Face detection
CN111737706B (zh) * 2020-05-11 2023-01-06 华南理工大学 一种生物特征隐私保护的前端人像加密与识别方法

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5533127A (en) * 1994-03-18 1996-07-02 Canon Information Systems, Inc. Encryption system
CN108776790A (zh) * 2018-06-06 2018-11-09 海南大学 云环境下基于神经网络的加密人脸识别方法
CN109657483A (zh) * 2018-11-27 2019-04-19 哈尔滨工业大学（深圳） 一种图像加密方法和系统
CN110633650A (zh) * 2019-08-22 2019-12-31 首都师范大学 基于隐私保护的卷积神经网络人脸识别方法及装置
CN111967033A (zh) * 2020-08-28 2020-11-20 深圳康佳电子科技有限公司 基于人脸识别的图片加密方法、装置、终端及存储介质
CN112949576A (zh) * 2021-03-29 2021-06-11 北京京东方技术开发有限公司 姿态估计方法、装置、设备及存储介质
CN113723334A (zh) * 2021-09-07 2021-11-30 合肥远康信息技术有限公司 一种金融联网综合安防系统
CN113807258A (zh) * 2021-09-17 2021-12-17 南京工业大学 一种基于神经网络与dct变换的加密人脸识别方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117998025A (zh) * 2024-04-07 2024-05-07 山东青橙数字科技有限公司 一种基于位阶计量运算的关键信息识别图像加密方法
CN117998025B (zh) * 2024-04-07 2024-06-11 山东青橙数字科技有限公司 一种基于位阶计量运算的关键信息识别图像加密方法

Also Published As

Publication number Publication date
CN117015953B (zh) 2024-07-09
CN117015953A (zh) 2023-11-07

Similar Documents

Publication Publication Date Title
Emad et al. A secure image steganography algorithm based on least significant bit and integer wavelet transform
Uludag et al. Securing fingerprint template: Fuzzy vault with helper data
Chen et al. Biometric based cryptographic key generation from faces
Gaddam et al. Efficient Cancelable Biometric Key Generation Scheme for Cryptography.
CN111723395B (zh) 一种人像生物特征隐私保护与解密方法
Olaniyi et al. Design of secure electronic voting system using fingerprint biometrics and crypto-watermarking approach
Yu et al. Privacy protecting visual processing for secure video surveillance
WO2024007095A1 (zh) 一种门禁系统人脸数据的安全加密方法和系统
Panchal et al. Comparable features and same cryptography key generation using biometric fingerprint image
Li et al. Riddle: Reversible and diversified de-identification with latent encryptor
CN104009973A (zh) 一种基于集合多项式变换与调和的指纹细节信息隐藏与恢复方法
Pic et al. Remote KYC: Attacks and counter-measures
CN112580011B (zh) 一种面向生物特征隐私保护的人像加解密系统
Balakumar et al. Secure biometric key generation scheme for cryptography using combined biometric features of fingerprint and iris
Lin et al. A Survey of DeepFakes Generation and Detection
CN110580405B (zh) 一种基于社交网络的图片版权保护系统及方法
Sowkarthika et al. Securing iris and fingerprint templates using fuzzy vault and symmetric algorithm
Nandini et al. Efficient cryptographic key generation from fingerprint using symmetric hash functions
Peethala et al. Integrating Biometric Cryptosystem with steganography for authentication
Meenakshi et al. Security analysis of hardened retina based fuzzy vault
CN113190858B (zh) 一种基于隐私保护的图像处理方法、系统、介质和设备
Maurya et al. A Survey on Cancellable Biometrics Techniques
Adamović et al. Information analysis of iris biometrics for the needs of cryptology key extraction
Patil et al. Secret sharing based secure authentication system
CN105719224A (zh) 基于对偶树复小波变换的压缩感知的生物特征图像加密方法

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 202280018231.8

Country of ref document: CN

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22949692

Country of ref document: EP

Kind code of ref document: A1