CN110795716A - Identity authentication method based on CNN, user equipment, storage medium and device - Google Patents
Identity authentication method based on CNN, user equipment, storage medium and device Download PDFInfo
- Publication number
- CN110795716A CN110795716A CN201910998390.9A CN201910998390A CN110795716A CN 110795716 A CN110795716 A CN 110795716A CN 201910998390 A CN201910998390 A CN 201910998390A CN 110795716 A CN110795716 A CN 110795716A
- Authority
- CN
- China
- Prior art keywords
- cnn
- behavior
- preset
- user
- training
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/316—User authentication by observing the pattern of computer usage, e.g. typical user behaviour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Social Psychology (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses a CNN-based identity authentication method, user equipment, a storage medium and a device. In the invention, biological behavior data is collected, the collected biological behavior data is preprocessed to obtain a training sample, a preset convolutional neural network model is trained based on the training sample, so that a man-machine distinguishing model is constructed, the man-machine distinguishing model is used for verifying the current biological behavior data of a user and the man-machine distinguishing model, and a verification result is output according to a judgment result. According to the technical scheme, the user behavior is verified through the man-machine distinguishing model obtained by presetting the convolutional neural network model, complex verification steps and operation are not needed, and the verification speed is greatly improved.
Description
Technical Field
The present invention relates to the internet field, and in particular, to a CNN-based authentication method, user equipment, storage medium, and apparatus.
Background
The verification code is a reverse Turing test and is used for man-machine distinguishing and blocking machine interaction requests. The traditional identifying code is a character type, and machine recognition is resisted by deforming, distorting and increasing interference on characters in a picture.
With the continuous development of technology, the success rate of recognizing traditional verification codes by means of OCR (Optical Character Recognition), machine learning and the like reaches up to 99%, and in order to resist automatic image Recognition, the traditional verification codes have to become more and more complex, and a user needs to spend long time and complicated verification operation during verification, so that the authentication efficiency is low.
Therefore, the technical problem of low authentication efficiency exists in the current webpage login authentication.
The above is only for the purpose of assisting understanding of the technical aspects of the present invention, and does not represent an admission that the above is prior art.
Disclosure of Invention
The invention mainly aims to provide a CNN-based identity authentication method, user equipment, a storage medium and a device, and aims to solve the technical problem of low authentication efficiency in webpage login authentication in the prior art.
In order to achieve the above object, the present invention provides an identity authentication method based on CNN, which comprises the following steps:
collecting biological behavior data;
preprocessing the collected biological behavior data to obtain a training sample;
training a preset convolutional neural network model based on the training sample to obtain a man-machine distinguishing model;
verifying the current biological behavior data of the user according to the man-machine distinguishing model to obtain a behavior similarity value;
judging whether the user behavior is normal or not according to the behavior similarity value and a preset verification threshold;
and outputting a verification result according to the judgment result.
Preferably, the collecting biological behavior data comprises:
acquiring mouse activity data of a user through the buried points;
and extracting a characteristic sample set from the mouse activity data, and taking the characteristic sample set as biological behavior data.
Preferably, the preprocessing the collected biological behavior data to obtain a training sample includes:
and extracting a mouse behavior feature vector with a preset dimension from the feature sample set, and taking the feature vector with the preset dimension as a training sample.
Preferably, the preset dimensions include a frequency subsection, a stationary event duty cycle, a moving time frequency, a single click time interval, a double click time interval, and an average moving speed.
Preferably, the method for training the preset convolutional neural network model based on the training sample to obtain the man-machine distinguishing model comprises the following steps:
setting a cost function;
inputting the training sample into a preset convolutional neural network model;
and training the convolutional neural network model by adjusting the cost function to obtain a man-machine distinguishing model.
Preferably, the pooling layer of the convolutional neural network model adopts a maximum pooling layer, the connection bias is initialized to zero, and the weight initialization is initialized by adopting a gaussian function.
Preferably, the method is characterized in that whether the user behavior is normal is judged according to the behavior similarity value and a preset verification threshold:
when the behavior similarity value is smaller than a preset verification threshold value, judging that the user behavior is abnormal;
and when the behavior similarity value is larger than a preset verification threshold value, judging that the user behavior is normal.
In order to achieve the above object, the present invention further provides a user equipment, where the user equipment includes: the identity authentication method comprises the following steps of a memory, a processor and a CNN-based identity authentication program stored on the memory and capable of running on the processor, wherein the steps of the CNN-based identity authentication method are realized when the CNN-based identity authentication program is executed by the processor.
In order to achieve the above object, the present invention further provides a storage medium, where a CNN-based authentication program is stored, and the CNN-based authentication program implements the steps of the CNN-based authentication method when executed by a processor.
In order to achieve the above object, the present invention further provides an authentication apparatus based on CNN, including:
the acquisition module is used for acquiring biological behavior data;
the data processing module is used for preprocessing the collected biological behavior data to obtain a training sample;
the training module is used for training a preset convolutional neural network model based on the training sample to obtain a man-machine distinguishing model;
the verification module is used for verifying the current biological behavior data of the user according to the man-machine distinguishing model to obtain a behavior similarity value;
the judging module is used for judging whether the user behavior is normal or not according to the behavior similarity value and a preset verification threshold;
and the output module is used for outputting the verification result according to the judgment result.
According to the technical scheme, the biological behavior data are collected, the collected biological behavior data are preprocessed to obtain training samples, a preset convolutional neural network model is trained based on the training samples, a man-machine distinguishing model is built, the man-machine distinguishing model is used for verifying the current biological behavior data of a user and the man-machine distinguishing model, and a verification result is output according to the judgment result. According to the technical scheme, the user behavior is verified through the man-machine distinguishing model obtained by presetting the convolutional neural network model, complex verification steps and operation are not needed, and the verification speed is greatly improved.
Drawings
FIG. 1 is a schematic diagram of a user equipment architecture of a hardware operating environment according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a CNN-based authentication method according to a first embodiment of the present invention;
FIG. 3 is a detailed flowchart of step S10 in FIG. 1;
FIG. 4 is a detailed flowchart of step S30 in FIG. 1;
fig. 5 is a functional block diagram of a CNN-based authentication apparatus according to a first embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As shown in fig. 1, the user equipment may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may comprise a Display screen (Display), an input unit such as keys, and the optional user interface 1003 may also comprise a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface). The memory 1005 may be a high-speed RAM memory or a non-volatile memory (e.g., a magnetic disk memory). The memory 1005 may alternatively be a storage device separate from the processor 1001.
Those skilled in the art will appreciate that the user equipment configuration shown in fig. 1 does not constitute a limitation of the user equipment and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components.
As shown in fig. 1, a memory 1005, which is a storage medium, may include therein a language detection program of an operating system, a network communication module, a user interface module, and a terminal.
In the user equipment shown in fig. 1, the network interface 1004 is mainly used for connecting an external network and performing data communication with other network equipment; the user interface 1003 is mainly used for connecting a user terminal and performing data communication with the terminal; the user equipment calls the building recommendation program stored in the memory 1005 through the processor 1001 and executes the language detection method of the terminal provided by the embodiment of the invention.
The user equipment can be electronic equipment such as a personal computer or a smart phone.
In the user equipment shown in fig. 1, the network interface 1004 is mainly used for connecting a terminal and communicating data with the terminal; and processor 1001 may be configured to invoke a CNN-based authentication procedure stored in memory 1003 and perform the following operations:
collecting biological behavior data;
preprocessing the collected biological behavior data to obtain a training sample;
training a preset convolutional neural network model based on the training sample to obtain a man-machine distinguishing model;
verifying the current biological behavior data of the user according to the man-machine distinguishing model to obtain a behavior similarity value;
judging whether the user behavior is normal or not according to the behavior similarity value and a preset verification threshold;
and outputting a verification result according to the judgment result.
Preferably, the collecting biological behavior data comprises:
acquiring mouse activity data of a user through the buried points;
and extracting a characteristic sample set from the mouse activity data, and taking the characteristic sample set as biological behavior data.
Preferably, the preprocessing the collected biological behavior data to obtain a training sample includes:
and extracting a mouse behavior feature vector with a preset dimension from the feature sample set, and taking the feature vector with the preset dimension as a training sample.
Preferably, the preset dimensions include a frequency subsection, a stationary event duty cycle, a moving time frequency, a single click time interval, a double click time interval, and an average moving speed.
Preferably, the method for training the preset convolutional neural network model based on the training sample to obtain the man-machine distinguishing model comprises the following steps:
setting a cost function;
inputting the training sample into a preset convolutional neural network model;
and training the convolutional neural network model by adjusting the cost function to obtain a man-machine distinguishing model.
Preferably, the pooling layer of the convolutional neural network model adopts a maximum pooling layer, the connection bias is initialized to zero, and the weight initialization is initialized by adopting a gaussian function.
Preferably, the method is characterized in that whether the user behavior is normal is judged according to the behavior similarity value and a preset verification threshold:
when the behavior similarity value is smaller than a preset verification threshold value, judging that the user behavior is abnormal;
and when the behavior similarity value is larger than a preset verification threshold value, judging that the user behavior is normal.
Based on the above hardware structure, the embodiment of the identity authentication method based on the CNN of the present invention is provided.
Referring to fig. 2, fig. 2 is a schematic flow chart of a CNN-based authentication method according to a first embodiment of the present invention.
In a first embodiment, the CNN-based authentication method includes the steps of:
step S10: collecting biological behavior data;
it is worth mentioning that the biological behavior data includes data generated by a keyboard, an iris, a fingerprint, a mouse, and the like. In this embodiment, data generated when the mouse is moved is used.
It should be noted that different users, due to different habits, have significant differences in the way they operate the mouse, such as the mouse single-double click time, frequency distribution, duty cycle of stationary events, moving time frequency, and average moving speed. And extracting the characteristics of the mouse operation modes to obtain the biological behavior data of the mouse.
Step S20: preprocessing the collected biological behavior data to obtain a training sample;
and classifying the obtained biological behavior data to obtain characteristic samples representing the aspects of single and double click time, frequency distribution, duty ratio of static events, moving time frequency, average moving speed and the like of the mouse.
Step S30: training a preset convolutional neural network model based on the training sample to obtain a man-machine distinguishing model;
convolutional Neural Networks (CNN) are a class of feed forward Neural Networks (fed forward Networks) that contain convolution computations and have a deep structure. Convolutional neural networks typically include an input layer, a hidden layer, and an output layer.
Wherein the input layer of the convolutional neural network can process multidimensional data. Hidden layers of the cumulative neural network comprise convolution layers, pooling layers and full-connection layers, which are 3 types of common structures. The function of the convolution layer is to perform feature extraction on input data, and the convolution layer internally includes a plurality of convolution kernels. After the feature extraction is performed on the convolutional layer, the output feature map is transmitted to the pooling layer for feature selection and information filtering. The pooling layer contains a pre-set pooling function whose function is to replace the result of a single point in the feature map with the feature map statistics of its neighboring regions. The step of selecting the pooling area by the pooling layer is the same as the step of scanning the characteristic diagram by the convolution kernel, and the pooling size, the step length and the filling are controlled. Upstream of the output layer in a convolutional neural network is typically a fully-connected layer. In the object detection problem, the output layer may be designed to output the center coordinates, size, and classification of an object, and the like.
In this embodiment, in the training, the training sample is input to the input layer of the preset convolutional neural network model to train the hidden layer, so as to obtain the man-machine distinguishing model.
Step S40: verifying the current biological behavior data of the user according to the man-machine distinguishing model to obtain a behavior similarity value;
and inputting the current biological behavior data needing to be verified into the human-computer distinguishing model, and outputting a behavior similarity value according to the human-computer distinguishing model.
Step S50: judging whether the user behavior is normal or not according to the behavior similarity value and a preset verification threshold;
it is worth to be noted that a preset verification threshold is set according to the result of processing the human biological behavior data and the machine biological behavior data by the human-computer distinguishing model, and the accuracy of verification can be adjusted by adjusting the preset verification threshold.
Step S60: and outputting a verification result according to the judgment result.
And further, displaying the judged result in a display screen mode or other prompting modes.
According to the technical scheme, the user behavior is verified through the man-machine distinguishing model obtained by presetting the convolutional neural network model, complex verification steps and operation are not needed, and the verification speed is greatly improved.
Referring to fig. 3, in a first embodiment, the acquiring biological behavior data includes:
step S11: acquiring mouse activity data of a user through the buried points;
it is worth to be noted that data embedding is a privatized deployment data collection method. In the present embodiment, the mouse activity data is used as the biological behavior data.
Step S12: and extracting a characteristic sample set from the mouse activity data, and taking the characteristic sample set as biological behavior data.
In this embodiment, data related to frequency distribution, stationary event duty cycle, moving time frequency, click time interval, double click time interval, and average moving speed is extracted from the mouse activity data and used as a feature sample set.
In a first embodiment, the preprocessing the collected biological behavior data to obtain a training sample includes:
and extracting a mouse behavior feature vector with a preset dimension from the feature sample set, and taking the feature vector with the preset dimension as a training sample.
It should be noted that, according to different types of mouse activity data, the mouse activity data can be divided into eight dimensions, which are respectively frequency distribution, stationary event duty cycle, moving time frequency, click time interval, double click time interval and average moving speed. And taking the mouse behavior feature vector extracted according to the dimension as a training sample.
The preset dimensionalities comprise frequency distribution, static event duty ratio, moving time frequency, single-click time interval, double-click time interval and average moving speed.
Referring to fig. 4, in a first embodiment, the method for obtaining a human-computer distinguishing model by training a preset convolutional neural network model based on the training samples includes:
step S31: setting a cost function;
in this embodiment, the cost function is used for parameter evaluation, specifically, a cross entropy loss function is used as the cost function.
Step S31: inputting the training sample into a preset convolutional neural network model;
in this embodiment, the pooling layer of the convolutional neural network model is a maximum pooling layer, the connection bias is initialized to zero, and the weight initialization is initialized by using a gaussian function.
Step S31: and training the convolutional neural network model by adjusting the cost function to obtain a man-machine distinguishing model.
The current behavior similarity value can be obtained from subsequent biological behavior data through the man-machine distinguishing model.
Further, the step of judging whether the user behavior is normal according to the behavior similarity value and a preset verification threshold value is as follows:
when the behavior similarity value is smaller than a preset verification threshold value, judging that the user behavior is abnormal;
and when the behavior similarity value is larger than a preset verification threshold value, judging that the user behavior is normal.
It can be understood that when the behavior similarity value is smaller than the preset verification threshold, it indicates that the current user behavior is not the normal operation performed by a human but an illegal behavior generated by a machine, and at this time, the user behavior verification fails, and a warning prompt can be given, and a defensive measure can be taken in time.
Referring to fig. 5, based on the above CNN-based authentication method, the present invention further provides a CNN-based authentication apparatus, where the CNN-based authentication apparatus includes:
an acquisition module 100, configured to acquire biological behavior data;
it is worth mentioning that the biological behavior data includes data generated by a keyboard, an iris, a fingerprint, a mouse, and the like. In this embodiment, data generated when the mouse is moved is used.
It should be noted that different users, due to different habits, have significant differences in the way they operate the mouse, such as the mouse single-double click time, frequency distribution, duty cycle of stationary events, moving time frequency, and average moving speed. And extracting the characteristics of the mouse operation modes to obtain the biological behavior data of the mouse.
The data processing module 200 is configured to preprocess the collected biological behavior data to obtain a training sample;
and classifying the obtained biological behavior data to obtain characteristic samples representing the aspects of single and double click time, frequency distribution, duty ratio of static events, moving time frequency, average moving speed and the like of the mouse.
The training module 300 is used for training a preset convolutional neural network model based on the training sample to obtain a man-machine distinguishing model;
convolutional Neural Networks (CNN) are a class of feed forward Neural Networks (fed forward Networks) that contain convolution computations and have a deep structure. Convolutional neural networks typically include an input layer, a hidden layer, and an output layer.
Wherein the input layer of the convolutional neural network can process multidimensional data. Hidden layers of the cumulative neural network comprise convolution layers, pooling layers and full-connection layers, which are 3 types of common structures. The function of the convolution layer is to perform feature extraction on input data, and the convolution layer internally includes a plurality of convolution kernels. After the feature extraction is performed on the convolutional layer, the output feature map is transmitted to the pooling layer for feature selection and information filtering. The pooling layer contains a pre-set pooling function whose function is to replace the result of a single point in the feature map with the feature map statistics of its neighboring regions. The step of selecting the pooling area by the pooling layer is the same as the step of scanning the characteristic diagram by the convolution kernel, and the pooling size, the step length and the filling are controlled. Upstream of the output layer in a convolutional neural network is typically a fully-connected layer. In the object detection problem, the output layer may be designed to output the center coordinates, size, and classification of an object, and the like.
In this embodiment, in the training, the training sample is input to the input layer of the preset convolutional neural network model to train the hidden layer, so as to obtain the man-machine distinguishing model.
The verification module 400 is used for verifying the current biological behavior data of the user according to the man-machine distinguishing model to obtain a behavior similarity value;
and inputting the current biological behavior data needing to be verified into the human-computer distinguishing model, and outputting a behavior similarity value according to the human-computer distinguishing model.
The judging module 500 is configured to judge whether the user behavior is normal according to the behavior similarity value and a preset verification threshold;
it is worth to be noted that a preset verification threshold is set according to the result of processing the human biological behavior data and the machine biological behavior data by the human-computer distinguishing model, and the accuracy of verification can be adjusted by adjusting the preset verification threshold.
And an output module 600, configured to output a verification result according to the determination result.
And further, displaying the judged result in a display screen mode or other prompting modes.
Preferably, the acquisition module 100 is further configured to acquire mouse activity data of a user through a buried point; and extracting a characteristic sample set from the mouse activity data, and taking the characteristic sample set as biological behavior data.
It is worth to be noted that data embedding is a privatized deployment data collection method. In the present embodiment, the mouse activity data is used as the biological behavior data. In this embodiment, data related to frequency distribution, stationary event duty cycle, moving time frequency, click time interval, double click time interval, and average moving speed is extracted from the mouse activity data and used as a feature sample set.
Preferably, the data processing module 200 is further configured to extract a mouse behavior feature vector of a preset dimension from the feature sample set, and use the feature vector of the preset dimension as a training sample. It should be noted that, according to different types of mouse activity data, the mouse activity data can be divided into eight dimensions, which are respectively frequency distribution, stationary event duty cycle, moving time frequency, click time interval, double click time interval and average moving speed. And taking the mouse behavior feature vector extracted according to the dimension as a training sample.
Preferably, the preset dimensions include a frequency subsection, a stationary event duty cycle, a moving time frequency, a single click time interval, a double click time interval, and an average moving speed.
Preferably, the training module 300 is configured to set a cost function; inputting the training sample into a preset convolutional neural network model; and training the convolutional neural network model by adjusting the cost function to obtain a man-machine distinguishing model. In this embodiment, the cost function is used for parameter evaluation, specifically, a cross entropy loss function is used as the cost function. The pooling layer of the convolutional neural network model adopts a maximum pooling layer, connection bias is initialized to be zero, and weight initialization is initialized by adopting a Gaussian function. The current behavior similarity value can be obtained from the subsequent biological behavior data through the man-machine distinguishing model
The pooling layer of the convolutional neural network model adopts a maximum pooling layer, connection bias is initialized to be zero, and weight initialization is initialized by adopting a Gaussian function.
Preferably, the determining module 500 is configured to determine that the user behavior is abnormal when the behavior similarity value is smaller than a preset verification threshold; and when the behavior similarity value is larger than a preset verification threshold value, judging that the user behavior is normal.
It can be understood that when the behavior similarity value is smaller than the preset verification threshold, it indicates that the current user behavior is not the normal operation performed by a human but an illegal behavior generated by a machine, and at this time, the user behavior verification fails, and a warning prompt can be given, and a defensive measure can be taken in time.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments. The use of the words first, second, third, etc. do not denote any order, but rather the words first, second, etc. are to be interpreted as names.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. An identity authentication method based on CNN, characterized in that, the identity authentication method based on CNN includes the following steps:
collecting biological behavior data;
preprocessing the collected biological behavior data to obtain a training sample;
training a preset convolutional neural network model based on the training sample to obtain a man-machine distinguishing model;
verifying the current biological behavior data of the user according to the man-machine distinguishing model to obtain a behavior similarity value;
judging whether the user behavior is normal or not according to the behavior similarity value and a preset verification threshold;
and outputting a verification result according to the judgment result.
2. The CNN-based authentication method of claim 1, wherein said collecting biological behavior data comprises:
acquiring mouse activity data of a user through the buried points;
and extracting a characteristic sample set from the mouse activity data, and taking the characteristic sample set as biological behavior data.
3. The CNN-based authentication method according to claim 2, wherein said preprocessing the collected biological behavior data to obtain a training sample comprises:
and extracting a mouse behavior feature vector with a preset dimension from the feature sample set, and taking the feature vector with the preset dimension as a training sample.
4. The CNN-based authentication method of claim 3, wherein the preset dimensions include frequency division, stationary event duty cycle, moving time frequency, single click time interval, double click time interval and average moving speed.
5. The CNN-based identity authentication method of claim 1, wherein the method for obtaining the human-machine distinction model by training the preset convolutional neural network model based on the training samples comprises:
setting a cost function;
inputting the training sample into a preset convolutional neural network model;
and training the convolutional neural network model by adjusting the cost function to obtain a man-machine distinguishing model.
6. The CNN-based authentication method of claim 1, wherein the pooling layer of the convolutional neural network model employs a maximum pooling layer, connection bias is initialized to zero, and weight initialization is initialized using a gaussian function.
7. The CNN-based identity verification method of any one of claims 1 to 6, wherein said determining whether the user behavior is normal according to the behavior similarity value and a preset verification threshold comprises:
when the behavior similarity value is smaller than a preset verification threshold value, judging that the user behavior is abnormal;
and when the behavior similarity value is larger than a preset verification threshold value, judging that the user behavior is normal.
8. A user equipment, the user equipment comprising: memory, processor and a CNN-based authentication program stored on the memory and executable on the processor, the CNN-based authentication program when executed by the processor implementing the steps of the CNN-based authentication method according to any one of claims 1 to 7.
9. A storage medium having stored thereon a CNN-based authentication program, which when executed by a processor implements the steps of the CNN-based authentication method according to any one of claims 1 to 7.
10. A CNN-based authentication apparatus, comprising:
the acquisition module is used for acquiring biological behavior data;
the data processing module is used for preprocessing the collected biological behavior data to obtain a training sample;
the training module is used for training a preset convolutional neural network model based on the training sample to obtain a man-machine distinguishing model;
the verification module is used for verifying the current biological behavior data of the user according to the man-machine distinguishing model to obtain a behavior similarity value;
the judging module is used for judging whether the user behavior is normal or not according to the behavior similarity value and a preset verification threshold;
and the output module is used for outputting the verification result according to the judgment result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910998390.9A CN110795716A (en) | 2019-10-22 | 2019-10-22 | Identity authentication method based on CNN, user equipment, storage medium and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910998390.9A CN110795716A (en) | 2019-10-22 | 2019-10-22 | Identity authentication method based on CNN, user equipment, storage medium and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110795716A true CN110795716A (en) | 2020-02-14 |
Family
ID=69439452
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910998390.9A Pending CN110795716A (en) | 2019-10-22 | 2019-10-22 | Identity authentication method based on CNN, user equipment, storage medium and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110795716A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112632494A (en) * | 2020-12-22 | 2021-04-09 | 江苏通付盾科技有限公司 | Mobile application identity verification method and device based on time sequence model |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107766852A (en) * | 2017-12-06 | 2018-03-06 | 电子科技大学 | A kind of man-machine mouse track detection method based on convolutional neural networks |
| CN107911338A (en) * | 2017-10-13 | 2018-04-13 | 深圳市迅雷网络技术有限公司 | A kind of data verification method, relevant device and system |
| US20180225516A1 (en) * | 2017-02-06 | 2018-08-09 | Brown University | Method and system for automated behavior classification of test subjects |
| CN109413023A (en) * | 2018-08-24 | 2019-03-01 | 阿里巴巴集团控股有限公司 | The training of machine recognition model and machine identification method, device, electronic equipment |
-
2019
- 2019-10-22 CN CN201910998390.9A patent/CN110795716A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180225516A1 (en) * | 2017-02-06 | 2018-08-09 | Brown University | Method and system for automated behavior classification of test subjects |
| CN107911338A (en) * | 2017-10-13 | 2018-04-13 | 深圳市迅雷网络技术有限公司 | A kind of data verification method, relevant device and system |
| CN107766852A (en) * | 2017-12-06 | 2018-03-06 | 电子科技大学 | A kind of man-machine mouse track detection method based on convolutional neural networks |
| CN109413023A (en) * | 2018-08-24 | 2019-03-01 | 阿里巴巴集团控股有限公司 | The training of machine recognition model and machine identification method, device, electronic equipment |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112632494A (en) * | 2020-12-22 | 2021-04-09 | 江苏通付盾科技有限公司 | Mobile application identity verification method and device based on time sequence model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108229130B (en) | Verification method and device | |
| US10445562B2 (en) | AU feature recognition method and device, and storage medium | |
| CN106951832B (en) | Verification method and device based on handwritten character recognition | |
| CN111191201B (en) | User identification method, device, equipment and storage medium based on data embedded point | |
| CN109409349B (en) | Credit certificate authentication method, credit certificate authentication device, credit certificate authentication terminal and computer readable storage medium | |
| CN109784181A (en) | Picture watermark recognition methods, device, equipment and computer readable storage medium | |
| CN109194689B (en) | Abnormal behavior recognition method, device, server and storage medium | |
| CN109740040B (en) | Verification code identification method, device, storage medium and computer equipment | |
| CN112231668A (en) | User identity authentication method based on keystroke behavior, electronic equipment and storage medium | |
| CN112001362A (en) | Image analysis method, image analysis device and image analysis system | |
| CN111881901A (en) | Screenshot content detection method and device and computer-readable storage medium | |
| CN102592142A (en) | Computer-system-based handwritten signature stability evaluation method | |
| CN112307994A (en) | Obstacle identification method based on sweeper, electronic device and storage medium | |
| CN111553241A (en) | Method, device and equipment for rejecting mismatching points of palm print and storage medium | |
| CN110717407A (en) | Human face recognition method, device and storage medium based on lip language password | |
| CN110737687A (en) | Data query method, device, equipment and storage medium | |
| CN110781467A (en) | Abnormal business data analysis method, device, equipment and storage medium | |
| CN110795716A (en) | Identity authentication method based on CNN, user equipment, storage medium and device | |
| CN111797395A (en) | Malicious code visualization and variety detection method, device, equipment and storage medium | |
| CN111797396A (en) | Malicious code visualization and variety detection method, device, equipment and storage medium | |
| WO2020224127A1 (en) | Video stream capturing method and apparatus, and storage medium | |
| CN110795705B (en) | Track data processing method, device and equipment and storage medium | |
| CN111797397B (en) | Malicious code visualization and variant detection method, device and storage medium | |
| CN114579965A (en) | Malicious code detection method and device and computer readable storage medium | |
| CN111930826A (en) | Order generation method and system of software interface |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200214 |