CN113807213B - Vehicle owner identity pre-authentication method based on behavior characteristics - Google Patents
Vehicle owner identity pre-authentication method based on behavior characteristics Download PDFInfo
- Publication number
- CN113807213B CN113807213B CN202111012376.0A CN202111012376A CN113807213B CN 113807213 B CN113807213 B CN 113807213B CN 202111012376 A CN202111012376 A CN 202111012376A CN 113807213 B CN113807213 B CN 113807213B
- Authority
- CN
- China
- Prior art keywords
- authentication
- pressure data
- user
- model
- similarity
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 239000013074 reference sample Substances 0.000 claims abstract description 19
- 238000007781 pre-processing Methods 0.000 claims abstract description 15
- 239000000523 sample Substances 0.000 claims abstract description 13
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 6
- 238000011176 pooling Methods 0.000 claims description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 5
- 230000003213 activating effect Effects 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 3
- 238000002372 labelling Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 3
- 239000013598 vector Substances 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Lock And Its Accessories (AREA)
Abstract
The invention discloses a vehicle owner identity pre-authentication method based on behavior characteristics, which specifically comprises the following steps: collecting pressure data of different volunteers when the doors of the vehicle are opened by using a pressure sensor, and carrying out manual labeling; preprocessing the pressure data, and training the authentication model by utilizing the preprocessed pressure data to obtain a trained authentication model; collecting pressure data of legal users of the vehicle when the legal users open the vehicle door for a plurality of times, and preprocessing the pressure data to be used as a reference sample set of the legal users; collecting pressure data of an unknown user when the door is opened, and preprocessing the pressure data to be used as an authentication sample; and comparing the similarity of the authentication sample with each reference sample in the reference sample set by using the trained authentication model, and pre-authenticating, opening the vehicle door if the pre-authentication is passed, otherwise, unlocking by using a key. The method and the device utilize the action characteristics of the user in the process of opening the door to pre-authenticate the legality of the user identity, and improve the user experience and the safety.
Description
Technical Field
The invention relates to a vehicle owner identity pre-authentication method based on behavior characteristics, and belongs to the technical field of identity authentication.
Background
In the traffic field, the identity authentication method for the vehicle owner mainly uses a key to unlock the vehicle, uses a wireless vehicle key to emit radio waves to unlock the vehicle remotely, uses a keyless entry technology to unlock the vehicle, and the like. However, these techniques have drawbacks: basically, all unlocking modes need to carry additional unlocking equipment, a car key is easy to copy by an attacker, remote unlocking and keyless entry are threatened by replay attack, and meanwhile, the keyless entry system has the defects of signature and certificate and is easy to attack. Therefore, in order to effectively authenticate the identity of the vehicle owner and improve the user experience and safety, a new identity authentication method is required to be provided.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: the vehicle owner identity pre-authentication method based on the behavior features can effectively pre-authenticate the identity of the vehicle owner so as to improve user experience and safety.
The invention adopts the following technical scheme for solving the technical problems:
The method comprises model training stage, user registration stage and authentication stage, wherein,
The model training phase comprises the following steps:
step 1, collecting pressure data of different volunteers when the doors of the volunteers are opened by using a pressure sensor, and manually marking the pressure data to mark the sources of the pressure data;
Step 2, preprocessing the pressure data, and training the authentication model by utilizing the preprocessed pressure data to obtain a trained authentication model;
the user registration phase includes the steps of:
Step 3, collecting pressure data of a legal user of the vehicle when the door is opened for a plurality of times by using a pressure sensor, and preprocessing the pressure data to be used as a reference sample set of the legal user;
the authentication phase comprises the following steps:
Step 4, collecting pressure data of an unknown user when the door is opened by using a pressure sensor, and preprocessing the pressure data to be used as an authentication sample;
and 5, comparing the similarity of the authentication sample with each reference sample in the reference sample set by using the trained authentication model, pre-authenticating, opening a vehicle door if the pre-authentication is passed, and unlocking by using a key if the pre-authentication is failed.
As a preferable mode of the present invention, the pressure sensor is a film type pressure sensor and is mounted on the inner side of the door handle.
As a preferred embodiment of the present invention, the preprocessing is specifically 0-1 standard normalization of the pressure data.
As a preferred scheme of the invention, the authentication model is a twin neural network and is composed of two feature extraction networks with the same parameters and shared by the same structure, the feature extraction network is formed by sequentially connecting a convolutional neural network and a pyramid pooling layer, the convolutional neural network comprises a first convolutional layer, a first activating layer, a second convolutional layer and a second activating layer which are sequentially connected, and the layer number of the pyramid pooling layer is 8;
In the authentication model training process, a contrast loss function is used for measuring the training of the model, and the contrast loss function is as follows:
Wherein L (w, (Y, S i,Sj)) represents a contrast loss function, w is a model parameter, E w(Si,Sj) is a similarity coefficient, Y is a label of the training set, and when two samples are samples of the same user, y=0; when two samples are samples of different users, y=1, m is the distance interval, and S i,Sj is any two samples.
As a preferable scheme of the invention, the specific process of the step 5 is as follows:
step 51, calculating the similarity between the authentication sample and each reference sample in the reference sample set by using the trained authentication model to obtain a similarity set;
Step 52, pre-authenticating an authentication sample of an unknown user by using an authentication method based on a voting idea, comparing each similarity in a similarity set with a threshold value T D, voting a reference sample which is larger than the threshold value T D and is considered to be corresponding to the similarity as authentication failure, and voting which is smaller than or equal to the threshold value as authentication success, thereby obtaining a voting set C, wherein T D is 1.2;
Step 53, counting votes in the voting set C, if the proportion of votes which are successfully authenticated to the total number of votes is greater than a threshold T C, the authentication sample is successfully pre-authenticated, and the door is opened; otherwise, the pre-authentication fails, requiring the user to unlock with a key, where T C is 0.8.
Compared with the prior art, the technical scheme provided by the invention has the following technical effects:
the invention provides a vehicle owner identity pre-authentication method based on behavior characteristics. Compared with the traditional vehicle owner authentication means, the user experience and the safety are improved.
Drawings
FIG. 1 is an overall architecture diagram of a vehicle owner identity pre-authentication method based on behavioral characteristics of the present invention.
Fig. 2 is a schematic diagram of an authentication model based on a twin neural network in the present invention.
FIG. 3 is a schematic diagram of a convolutional layer of a twin neural network.
FIG. 4 is a schematic diagram of a pyramidal pooling layer of a twin neural network.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings. The embodiments described below by referring to the drawings are exemplary only for explaining the present invention and are not to be construed as limiting the present invention.
The invention collects pressure data when a user pulls the door handle by using the film pressure sensor, analyzes the behavior habit of the user and performs pre-authentication, if the user passes the authentication, the door is opened, otherwise, the door is unlocked by using a traditional key.
As shown in fig. 1, the overall architecture diagram of the vehicle owner identity pre-authentication method based on behavior features of the present invention is divided into three stages: the method comprises a model training stage, a user registration stage and an authentication stage, wherein the three stages comprise two common steps: a data acquisition step and a data preprocessing step; meanwhile, the model training stage further comprises training of an authentication model, the user registration stage further comprises storing of an authentication reference set of a legal user, and the authentication stage further comprises authenticating an unknown user by using the trained model.
And a data acquisition step: the method comprises the steps that a film type pressure sensor is additionally arranged on the inner side of a door handle, pressure change of the door handle when a user pulls the door handle is collected in real time, manual data marking is carried out, a data source is marked, and collected data are in a time sequence data format;
A data preprocessing step: carrying out 0-1 standard normalization on the pressure data in the data preprocessing step;
training an authentication model: before leaving the factory, pressure data when a plurality of volunteers open the car door for many times are collected in advance to be used as a training set, and the twin neural network is trained to obtain the capability of similarity calculation of time sequence data.
The structure diagram of the twin neural network is shown in fig. 2, and the twin neural network is composed of two feature extraction networks with the same parameters and shared by the two structures, wherein the feature extraction networks are composed of a one-dimensional convolutional neural network (fig. 3) and a pyramid pooling layer (fig. 4). The former is composed of two one-dimensional convolution layers and an activation layer, and the number of layers of the latter is 8, and the two layers are used for generating feature vectors with the same length so as to facilitate similarity calculation. And performing similarity calculation on the two feature vectors with the same length by using the L2 norm. In the training process, the contrast loss is used for measuring the training of the network, and the contrast loss function is as follows:
Wherein E w(Si,Sj) is a similarity coefficient, Y is a label of the training set, and when two samples are samples of the same user, y=0; when two samples are samples of different users, y=1, w is a model parameter, m is a distance interval, which is a super parameter, set to 2, and s i,Sj is any two samples.
Registering: in the registration stage, a legal user pulls the vehicle door for a plurality of times, the pressure sensor collects pressure data of the user, and after preprocessing and storage, a reference sample set of the legal user is formed.
Identity authentication: in the authentication stage, pressure data are acquired when a user pulls a vehicle door, authentication samples of the user are obtained after the pressure data are preprocessed, and similarity between the authentication samples and all reference samples in the reference sample set is calculated by using a trained authentication model, so that a similarity set is obtained. Authentication samples of the user are authenticated using a voting idea based authentication method. And comparing each similarity coefficient in the similarity set with a threshold T D, wherein the reference sample is considered to be voting as authentication failure when the similarity coefficient is larger than the threshold, and the reference sample is considered to be voting as authentication success when the similarity coefficient is smaller than or equal to the threshold, so as to obtain a voting set C, wherein T D is 1.2. The votes in the voting set C are counted, if the proportion of votes that succeed in authentication to the total number of votes is greater than a threshold T C, the authentication sample succeeds in authentication, otherwise the authentication fails, wherein T C is 0.8.
The above embodiments are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereto, and any modification made on the basis of the technical scheme according to the technical idea of the present invention falls within the protection scope of the present invention.
Claims (3)
1. A pre-authentication method for the identity of car owner based on behavior features is characterized by comprising a model training stage, a user registration stage and an authentication stage,
The model training phase comprises the following steps:
step 1, collecting pressure data of different volunteers when the doors of the volunteers are opened by using a pressure sensor, and manually marking the pressure data to mark the sources of the pressure data;
Step 2, preprocessing the pressure data, and training the authentication model by utilizing the preprocessed pressure data to obtain a trained authentication model;
The authentication model is a twin neural network and is composed of two feature extraction networks with the same parameters, wherein the feature extraction networks are formed by sequentially connecting a convolutional neural network and pyramid pooling layers, the convolutional neural network comprises a first convolutional layer, a first activating layer, a second convolutional layer and a second activating layer which are sequentially connected, and the number of layers of the pyramid pooling layers is 8;
In the authentication model training process, a contrast loss function is used for measuring the training of the model, and the contrast loss function is as follows:
Wherein L (w, (Y, S i,Sj)) represents a contrast loss function, w is a model parameter, E w(Si,Sj) is a similarity coefficient, Y is a label of the training set, and when two samples are samples of the same user, y=0; when two samples are samples of different users, y=1, m is the distance interval, and S i,Sj is any two samples;
the user registration phase includes the steps of:
Step 3, collecting pressure data of a legal user of the vehicle when the door is opened for a plurality of times by using a pressure sensor, and preprocessing the pressure data to be used as a reference sample set of the legal user;
the authentication phase comprises the following steps:
Step 4, collecting pressure data of an unknown user when the door is opened by using a pressure sensor, and preprocessing the pressure data to be used as an authentication sample;
Step 5, performing similarity comparison on the authentication sample and each reference sample in the reference sample set by using the trained authentication model, performing pre-authentication, opening a vehicle door if the pre-authentication is passed, and unlocking by using a key if the pre-authentication is failed; the specific process is as follows:
step 51, calculating the similarity between the authentication sample and each reference sample in the reference sample set by using the trained authentication model to obtain a similarity set;
Step 52, pre-authenticating an authentication sample of an unknown user by using an authentication method based on a voting idea, comparing each similarity in a similarity set with a threshold value T D, voting a reference sample which is larger than the threshold value T D and is considered to be corresponding to the similarity as authentication failure, and voting which is smaller than or equal to the threshold value as authentication success, thereby obtaining a voting set C, wherein T D is 1.2;
Step 53, counting votes in the voting set C, if the proportion of votes which are successfully authenticated to the total number of votes is greater than a threshold T C, the authentication sample is successfully pre-authenticated, and the door is opened; otherwise, the pre-authentication fails, requiring the user to unlock with a key, where T C is 0.8.
2. The vehicle owner identity pre-authentication method based on behavior characteristics according to claim 1, wherein the pressure sensor is a film type pressure sensor and is installed inside a vehicle door handle.
3. The vehicle owner identity pre-authentication method based on behavior characteristics according to claim 1, wherein the preprocessing is specifically 0-1 standard normalization of pressure data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111012376.0A CN113807213B (en) | 2021-08-31 | 2021-08-31 | Vehicle owner identity pre-authentication method based on behavior characteristics |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111012376.0A CN113807213B (en) | 2021-08-31 | 2021-08-31 | Vehicle owner identity pre-authentication method based on behavior characteristics |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113807213A CN113807213A (en) | 2021-12-17 |
CN113807213B true CN113807213B (en) | 2024-05-28 |
Family
ID=78942223
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111012376.0A Active CN113807213B (en) | 2021-08-31 | 2021-08-31 | Vehicle owner identity pre-authentication method based on behavior characteristics |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113807213B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110533631A (en) * | 2019-07-15 | 2019-12-03 | 西安电子科技大学 | SAR image change detection based on the twin network of pyramid pondization |
CN111371951A (en) * | 2020-03-03 | 2020-07-03 | 北京航空航天大学 | Smart phone user authentication method and system based on electromyographic signals and twin neural network |
CN112487374A (en) * | 2020-12-04 | 2021-03-12 | 重庆邮电大学 | Self-adaptive continuous identity authentication method and system based on touch screen interaction behavior |
-
2021
- 2021-08-31 CN CN202111012376.0A patent/CN113807213B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110533631A (en) * | 2019-07-15 | 2019-12-03 | 西安电子科技大学 | SAR image change detection based on the twin network of pyramid pondization |
CN111371951A (en) * | 2020-03-03 | 2020-07-03 | 北京航空航天大学 | Smart phone user authentication method and system based on electromyographic signals and twin neural network |
CN112487374A (en) * | 2020-12-04 | 2021-03-12 | 重庆邮电大学 | Self-adaptive continuous identity authentication method and system based on touch screen interaction behavior |
Also Published As
Publication number | Publication date |
---|---|
CN113807213A (en) | 2021-12-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN1972186B (en) | A mobile identity authentication system and its authentication method | |
US20050226475A1 (en) | Method of, and system for, accessing a home or dwelling | |
CN104700018B (en) | A kind of recognition methods for intelligent robot | |
CN102385766A (en) | Palmprint-based authentication unlocking method, terminal and system | |
CN108648310A (en) | A kind of face recognition door control system and its application process of double authentication | |
CN112734988A (en) | Forced unlocking judging and alarming system and method for intelligent door lock | |
US11611881B2 (en) | Integrated systems and methods for passive authentication | |
CN107444337A (en) | Automotive theft proof system with double authentication | |
CN107481736A (en) | A kind of vocal print identification authentication system and its certification and optimization method and system | |
CN108334828A (en) | A kind of vehicle-mounted 3D human face recognizing identity authentication systems | |
EP2517150A1 (en) | Method and system for generating a representation of a finger print minutiae information | |
CN113807213B (en) | Vehicle owner identity pre-authentication method based on behavior characteristics | |
CN109286499B (en) | Behavior feature-based presence authentication method | |
CN113241081B (en) | Far-field speaker authentication method and system based on gradient inversion layer | |
CN112566117B (en) | Vehicle node identity recognition method and device based on metric learning | |
CN109215189A (en) | Intelligent door lock open method and Verification System | |
CN206781701U (en) | A kind of identification key system of rent-a-car | |
CN117133073A (en) | Authority control method of intelligent door | |
CN113365273B (en) | Packet-level wireless equipment authentication method based on channel state information | |
CN115424378A (en) | Safety protection method and device of intelligent coded lock and related equipment | |
CN112669501B (en) | Access control method, device and computer readable storage medium | |
CN113660670B (en) | Radio frequency fingerprint-based wireless equipment identity authentication method and device | |
JP2004068411A (en) | Entering/leaving control device | |
Divan et al. | A fingerprint matching technique using minutiae based algorithm for voting system: A survey | |
CN110942536B (en) | Fingerprint identification unlocking system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |