Identity authentication method based on multimedia video information
By using a multimedia video information-based identity authentication method and employing centralized discretization technology for hash storage, the problem of existing text and graphic passwords being easily forgotten and vulnerable to attack is solved, thereby achieving enhanced security and a user-friendly identity authentication experience.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHANGHAI JIAOTONG UNIV
- Filing Date
- 2024-11-14
- Publication Date
- 2026-05-15
AI Technical Summary
Existing text-based and graphical password authentication schemes are easily forgotten, increasing the user's memory burden, and are vulnerable to dictionary attacks and social engineering attacks, resulting in insufficient security.
An identity authentication method based on multimedia video information is adopted, which generates identity authentication credentials by combining images and interactions, and uses centralized discretization technology for hash storage to expand the password space and improve security.
It significantly increases the size of the password space, simplifies user operations, reduces the false rejection rate, effectively resists dictionary attacks and social engineering attacks, and ensures user information security.
Smart Images
Figure CN122053095A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a technology in the field of information security, specifically an identity authentication method based on multimedia video information. Background Technology
[0002] Existing enhanced user password technologies have many limitations, such as being easily forgotten and increasing the user's memory burden. Graphical password authentication schemes also face potential risks of dictionary attacks or social engineering attacks. These limitations make these schemes less practical and hinder their widespread adoption as mature authentication solutions. Summary of the Invention
[0003] This invention addresses the problems of current mainstream identity authentication schemes such as text cryptography and graphic cryptography, as well as the shortcomings of existing technologies in terms of cryptographic space, which cannot meet the needs and are vulnerable to dictionary attacks and social engineering attacks, thus affecting security. It proposes an identity authentication method based on multimedia video information, which uses a hybrid approach of image and interaction as identity authentication credentials, significantly increasing the scale of the cryptographic space. At the same time, it utilizes a centralized discretization technique to hash and store user information, further enhancing the security of user information.
[0004] This invention is achieved through the following technical solution:
[0005] This invention relates to an identity authentication method based on multimedia video information, comprising:
[0006] Step 1, Registration Phase: The server collects and verifies the identity via video and the feature values of the selected location, performs hash value calculation using a central discretization technique, and generates a salted hash value authentication database.
[0007] The aforementioned center discretization technique refers to setting a point x on a one-dimensional straight line L with a continuous set of data points and discretizing the straight line L into segments of equal length. Point x falls on the center of the line segment containing it, that is, there is an equal tolerance r on both sides of point x. During the registration phase, point x is characterized by calculating the segment identifier i and the offset d, where the offset d is stored in plaintext and the segment identifier i is stored in a protected form as a salted hash value h(i,d).
[0008] The identifier Offset d = (xr) mod 2r, where: tolerance r is set according to system requirements or user preferences to determine system security and user availability.
[0009] The selected password during the authentication process is reconstructed using a central discretization technique. After reconstruction, the password can be subjected to the same salted hash calculation as the text password.
[0010] The registration phase specifically includes:
[0011] 1) The user selects a local video, selects the password at a specific time point, and obtains the password's location information (x, y), the video time point t, and the video feature value V.
[0012] The video feature value refers to the hash value calculation performed on the video file uploaded by the user to obtain a video feature value that uniquely corresponds to the user's identity.
[0013] The hash algorithm described uses, but is not limited to, various hash digest algorithms such as SHA2.
[0014] 2) Calculation d n = (nr) mod 2r, thus obtaining the hash value. h = hash(i x d x i y d y i t d t ) and local storage (d x d y d t ), where: n = x, y, t.
[0015] 3) The user sends the user ID, hash value h, and video feature value V to the server.
[0016] 4) The server stores the hash value h and the video feature value V in the database.
[0017] Step 2, Authentication Phase: The server compares the received video and selected location feature value information with the calculated salted hash value against the authentication database to obtain the authentication result. Specifically, this involves: during login calculation... That is, using the same offset d, calculate which segment of the line L contains the input point x′ to verify whether the input point x′ is acceptable. When i′=i, the hash value h(i′,d)=h(i,d) and the system accepts the input point x′; otherwise, i′≠i, that is, the input point x′ falls outside the set acceptable tolerance r, then the hash value h(i′,d)≠h(i,d) and the authentication fails. Where: i is the sequence index of the grid cell selected by the box during registration, and i′ is the sequence index of the grid cell selected by the box during login.
[0018] The authentication phase specifically includes:
[0019] a) The user selects a local video, selects the password at a specific time point, and obtains the password's location information (x′, y′), the video time point t′, and the video feature value V′.
[0020] b) Calculation Hash value h′ = hash(i′) x d x , i′ y d y , i′ t d t ), where: n = x, y, t.
[0021] c) The user sends the hash value h′ and the video feature value V′ to the server.
[0022] d) The server performs the comparison and returns the authentication result to the user.
[0023] This invention relates to a system for implementing the above method, comprising: a video feature value acquisition unit, a central discretization unit, a salted hash value calculation unit, and an identity information verification unit, wherein: the video feature value acquisition unit acquires feature values of user-uploaded videos and selected locations, and passes them to the central discretization unit; the central discretization unit performs central discretization processing on the feature values, and passes the discretized feature values to the salted hash value calculation unit; the salted hash value calculation unit calculates the hash value, and sends the hash value to the identity information verification unit for storage or comparison to obtain the registration or authentication result. Technical effect
[0024] This invention uses user-uploaded videos to generate authentication information, transforming user-uploaded videos and selected areas into authentication information, greatly expanding the password space. It generates authentication information in a user-friendly manner, making the login process simple, convenient, easy for users to understand and remember, reducing the system's false rejection rate, and providing good resistance to common attacks such as dictionary attacks and social engineering attacks. Attached Figure Description
[0025] Figure 1 This is a flowchart of the registration phase of the present invention;
[0026] Figure 2 This is a flowchart of the login phase of the present invention;
[0027] Figure 3 This is a schematic diagram of the central discretization process in the example;
[0028] Figure 4 This is a schematic diagram illustrating the effect of an example. Detailed Implementation
[0029] This embodiment relates to an identity authentication method based on multimedia video information. After processing the user-uploaded video and the selected password area, identity authentication information is obtained, specifically including:
[0030] like Figure 1As shown, the registration phase specifically includes:
[0031] Step 1: Display the user interface. The various components of the user interface are built using Python QT. After clicking the registration button, the user enters the main video playback interface. The password interception operation required in this example is performed on this interface. The user first clicks the "Select Video" button to select a local video. The selection operation filters the selectable filenames to prevent the user from selecting non-video files, thus affecting the normal function of the system or performing illegal operations.
[0032] Step 2: After the user successfully uploads the local video, load and play the uploaded video, and provide the user with relevant operation buttons for the video. Specifically, set up a video playback window on the page, with a length of 1280px and a width of 720px. At the bottom of the window, add buttons for pausing the video, capturing the password, and completing the capture. Provide a progress bar for video playback and display the current time and total video duration to allow users to quickly locate the video.
[0033] Step 3: After the user clicks the pause video button and the capture password button, a rectangle is captured as the password. Specifically, the mouse event in the playback window is rewritten to display the rectangle and obtain the coordinate position information (x0, y0, x1, y1), the video time point r, and the video feature value V.
[0034] Step 4: After the user confirms the selection, the system performs center discretization based on the acquired coordinate information (x0, y0, x1, y1), video time point t, and video feature value v. Here, (x0, y0) represents the starting point, and (x1, y1) represents the ending point. Calculate: starting point offset d0, starting point segment identifier i0, ending point offset d1, ending point segment identifier i1, and time point offset d. t Time node segment identifier i t At the same time, all offset parameters are stored in the local cache file without processing.
[0035] In this embodiment, the center discretization parameter is set to r = 80px. In specific applications, the value of r can be increased or decreased as needed to obtain different security levels and user experience levels.
[0036] Step 5: Calculate the salted hash value based on the calculated center discretization result. h=hash(d0,i0,d1,i1,d t i t The hash algorithm chosen is SHA2 or other hash digest algorithms.
[0037] Step 6: Send the hash value h and video feature value v to the server for storage. The registration process is now complete.
[0038] like Figure 2 As shown, the login phase specifically includes:
[0039] Steps a and b are the same as steps 1 and 2 in the registration phase.
[0040] Step c: The user captures the position captured during registration from the uploaded video, obtains the coordinate position information (x0′, y0′, x1′, y1′), the video time point t′, and the video feature value v′.
[0041] Step d: Based on the coordinate position information (x0′, y0′, x1′, y1′), video time point t′, and video feature value v′, perform center discretization processing. Calculate: the discretization label of the starting point i′0, the discretization label of the ending point i′1, and the discretization label of the time node i′ t .
[0042] Step e: Retrieve the offset parameters corresponding to the current user's cache file, namely the starting point offset d0, the ending point offset d1, and the time node offset d2. t Based on the calculated center discretization results, the salted hash value h′=hash(d0,i′0,d1,i′1,d t , i′ t ).
[0043] Step f: Send the hash value h′ and video feature value v′ to the server for comparison. If the comparison is successful, the login request is approved; otherwise, authentication fails.
[0044] Through specific practical experiments, under the specific environment setting of using a central discretization parameter r = 80px, users applied the above method for identity authentication, and the authentication experimental results are as follows: Figure 3 and Figure 4 As shown, the login success rate is 95.35%, the false acceptance rate is 0%, the average registration time is 15.12 seconds with a median of 12.50 seconds, and the average login time is 14.45 seconds with a median of 12.75 seconds. Furthermore, theoretical analysis shows that the password generated using this embodiment is significantly stronger than traditional text passwords and existing pattern passwords, and the password space is large enough to resist brute-force attacks. In addition, no user identity information is leaked, and it resists the threats of dictionary attacks and social engineering attacks.
[0045] Compared to existing technologies, this method is user-friendly, with shorter registration and login times, and easier-to-remember passwords. Compared to traditional text password authentication, users are more familiar with personal video authentication, resulting in a lower false rejection rate and making it easier for legitimate users to complete authentication. Furthermore, it significantly expands the password space, making it resistant to brute-force attacks while preserving the user's personal identity information and reducing the potential risks of dictionary attacks and social engineering attacks.
[0046] The above-described specific implementations can be partially adjusted by those skilled in the art in different ways without departing from the principles and purpose of the present invention. The scope of protection of the present invention is defined by the claims and is not limited to the above-described specific implementations. All implementation schemes within the scope of the claims are bound by the present invention.
Claims
1. An identity authentication method based on multimedia video information, characterized in that, During the registration phase, the server collects video footage and selected location feature values to verify identity, performs hash value calculations using a central discretization technique, and generates a salted hash value authentication database. During the verification phase, the server compares the received video and the feature value information of the selected position, as well as the calculated salted hash value, with the authentication database to obtain the authentication result.
2. The identity authentication method based on multimedia video information according to claim 1, characterized in that, The aforementioned center discretization technique refers to setting a point x on a one-dimensional straight line L with a continuous set of data points and discretizing the straight line L into segments of equal length, with point x falling at the center of the line segment containing it, that is, there is an equal tolerance r on both sides of point x; during the registration phase, the point is characterized by calculating the segment identifier i and the offset d, where: the offset d is stored in plaintext, and the segment identifier i is stored in a protected form as a salted hash value h(i,d).
3. The identity authentication method based on multimedia video information according to claim 2, characterized in that, The identifier Offset d = (xr) mod 2r, where: tolerance r is set according to system requirements or user preferences to determine system security and user availability.
4. The identity authentication method based on multimedia video information according to any one of claims 1-3, characterized in that, The aforementioned registration phase specifically includes: 1) The user selects a local video, selects the password at a specific time point, and obtains the password's location information (x, y), the video time point t, and the video feature value v; 2) Calculation d n = (nr)mod 2r, thus obtaining the hash value h = hash(i x d x i y d y i t d t ) and local storage (d x d y d t ), where: n = x, y, t; 3) The user sends their user ID, hash value h, and video feature value v to the server; 4) The server stores the hash value h and the video feature value v in the database.
5. The identity authentication method based on multimedia video information according to claim 1, characterized in that, The authentication phase mentioned above, during login calculation That is, using the same offset d, calculate which segment of the line L contains the input point x′ to verify whether the input point x′ is acceptable. When i′=i, the hash value h(i′,d)=h(i,d) and the system accepts the input point x′; otherwise, i′≠i′, that is, the input point x′ falls outside the set acceptable tolerance r, then the hash value h(i′,d)≠h(i,d) and the authentication fails. Where: i is the sequence index of the grid cell selected by the box during registration, and i′ is the sequence index of the grid cell selected by the box during login.
6. The identity authentication method based on multimedia video information according to claim 1 or 5, characterized in that, The authentication phase specifically includes: a) When a user selects a local video and frames the password at a specific time point, the location information of the password (x′, y′), the video time point, and the video feature value v′ are obtained. b) Calculation Hash value h′ = hash(i′) x d x , i′ y d y , i′ t d t ), where: n = x, y, t; c) The user sends the hash value h′ and the video feature value v′ to the server; d) The server performs the comparison and returns the authentication result to the user.
7. The identity authentication method based on multimedia video information according to claim 4 or 6, characterized in that, The video feature value refers to the hash value calculation performed on the video file uploaded by the user to obtain a video feature value that uniquely corresponds to the user's identity.
8. A system for implementing the identity authentication method based on multimedia video information according to any one of claims 1-7, characterized in that, include: The system comprises a video feature value acquisition unit, a central discretization unit, a salted hash value calculation unit, and an identity information verification unit. Specifically, the video feature value acquisition unit acquires the feature values of the user-uploaded video and the selected location, and passes them to the central discretization unit. The central discretization unit performs central discretization processing on the feature values and passes the discretized feature values to the salted hash value calculation unit. After calculating the hash value, the salted hash value calculation unit sends the hash value to the identity information verification unit for storage or comparison to obtain the registration or authentication result.