CN118551364A - Commercial password security assessment method and system based on deep learning - Google Patents
Commercial password security assessment method and system based on deep learning Download PDFInfo
- Publication number
- CN118551364A CN118551364A CN202411027828.6A CN202411027828A CN118551364A CN 118551364 A CN118551364 A CN 118551364A CN 202411027828 A CN202411027828 A CN 202411027828A CN 118551364 A CN118551364 A CN 118551364A
- Authority
- CN
- China
- Prior art keywords
- password
- commercial
- vulnerable
- evaluated
- model
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 96
- 238000013135 deep learning Methods 0.000 title claims abstract description 36
- 239000013598 vector Substances 0.000 claims abstract description 122
- 238000011156 evaluation Methods 0.000 claims abstract description 121
- 238000012545 processing Methods 0.000 claims abstract description 42
- 238000013210 evaluation model Methods 0.000 claims abstract description 37
- 238000006243 chemical reaction Methods 0.000 claims abstract description 12
- 230000006870 function Effects 0.000 claims description 36
- 230000009466 transformation Effects 0.000 claims description 19
- 238000012549 training Methods 0.000 claims description 17
- 238000003062 neural network model Methods 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 11
- 238000012216 screening Methods 0.000 claims description 5
- 238000011426 transformation method Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 230000001502 supplementing effect Effects 0.000 claims description 3
- 230000001131 transforming effect Effects 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 9
- 230000002093 peripheral effect Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000010606 normalization Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000001965 increasing effect Effects 0.000 description 3
- 238000003672 processing method Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 239000000945 filler Substances 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 239000000047 product Substances 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 238000011524 similarity measure Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000002688 persistence Effects 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Landscapes
- Storage Device Security (AREA)
Abstract
The application relates to the technical field of network security, in particular to a commercial password security assessment method and a commercial password security assessment system based on deep learning, wherein the method comprises the steps of obtaining a commercial password sample set, wherein the commercial password sample set comprises a vulnerable password set, a non-vulnerable password set, corresponding password strength and a vulnerability type corresponding to the vulnerable password set; performing consistency processing on the vulnerable cipher set and the non-vulnerable cipher set by using a cipher conversion method to generate vector representation; the method comprises the steps that a vector representation of a vulnerable password set and a vector representation of a non-vulnerable password set are used as input data, and corresponding password strength, whether a vulnerability exists or not and specific vulnerability types are used as output to train a model to obtain a commercial password evaluation model; obtaining a password to be evaluated, and performing consistency processing on the password to be evaluated by using a password position coding method to generate vector representation; and inputting the vector representation of the password to be evaluated into a commercial password evaluation model, and outputting an evaluation result of the password to be evaluated by the model. The method and the device can improve accuracy and efficiency of commercial password security assessment.
Description
Technical Field
The application relates to the technical field of network security, in particular to a commercial password security assessment method and system based on deep learning.
Background
In the current digital and informationized process, commercial passwords play a vital role as a basic means for guaranteeing data security. Commercial passwords are used not only for personal user account and data protection, but also for secure storage and transmission of confidential information and transaction data for enterprises. Therefore, ensuring the security of the commercial password is an important task in the daily operation of enterprises, and before the commercial password is used, the commercial password needs to be accurately evaluated, and the strength and the existence of a vulnerability of the commercial password are judged so as to ensure the security of the commercial password in use.
Security assessment of commercial passwords presents multiple challenges in practical applications. In the conventional evaluation method, enterprises often rely on internal teams or delegated to third party institutions to perform security evaluation of commercial passwords regularly, however, the conventional method is limited by professional ability and technical level of an evaluator, so that an evaluation result of the commercial passwords may have a problem of insufficient accuracy. Under the condition, the security of the commercial password application cannot be effectively ensured, so that the risk of information security holes is increased. In addition, the evaluation is performed by an internal team or a third party organization, so that timeliness of the evaluation is often caused, and the security evaluation work of the commercial passwords is difficult to complete rapidly, so that the evaluation efficiency of the commercial passwords is affected.
In summary, with the wide application and continuous remarkable achievement of the deep learning technology in various fields, how to apply the deep learning technology to the security evaluation of commercial passwords to solve the problems of insufficient accuracy and low efficiency of the security evaluation of commercial passwords in the traditional evaluation method is a challenge at present.
Disclosure of Invention
The application provides a commercial password security assessment method and a commercial password security assessment system based on deep learning, which can effectively improve the accuracy and efficiency of commercial password security assessment. The application provides the following technical scheme:
In a first aspect, the present application provides a method for evaluating security of commercial passwords based on deep learning, the method comprising:
Obtaining a commercial password sample set, wherein the commercial password sample set comprises a vulnerable password set, a non-vulnerable password set, password intensities corresponding to the vulnerable password set and the non-vulnerable password set and vulnerability types corresponding to the vulnerable password set;
Performing consistency processing on the vulnerable cipher set and the non-vulnerable cipher set by using a cipher conversion method based on a hash function, and generating vector representations of fixed lengths of the vulnerable cipher set and the non-vulnerable cipher set;
Training a preset neural network model by taking the vector representations of the fixed lengths of the vulnerable cipher set and the non-vulnerable cipher set as input data and taking the corresponding cipher strength, the existence of the vulnerability and the specific vulnerability type as output to obtain a commercial cipher evaluation model;
Obtaining a password to be evaluated, performing consistency processing on the password to be evaluated by using a password position coding method, and generating a vector representation of a fixed length of the password to be evaluated;
And inputting the vector representation with the fixed length of the password to be evaluated into the commercial password evaluation model, and outputting an evaluation result of the password to be evaluated by the commercial password evaluation model.
In a specific embodiment, the performing, using a hash function-based cryptographic transformation method, a consistency process on the vulnerable cryptographic set and the non-vulnerable cryptographic set, and generating a vector representation of the vulnerable cryptographic set and a fixed length of the non-vulnerable cryptographic set includes:
Setting a fixed length;
Converting the commercial password into a hash value through a hash function;
performing transformation operation on the hash value by using a transformation formula;
Intercepting the hash value after the transformation operation into a vector representation with a fixed length;
The vector representation is normalized.
In a specific embodiment, the converting the commercial password into the hash value by the hash function, and the transforming the hash value using the transformation formula includes:
all commercial passwords are converted to hash values using a SHA-256 hash function, which is shown below:
wherein, The commercial code is represented by a code of commerce,Representing the converted hash value;
Transforming the hash value using a transformation formula:
wherein, Representing a bitwise exclusive or operation,Representing the presentation to beThe two positions are shifted to the right,Representation fetchIs inverted by bit.
In a specific embodiment, the obtaining the password to be evaluated, performing consistency processing on the password to be evaluated by using a password position coding method, and generating the vector representation of the fixed length of the password to be evaluated includes:
If the length of the password to be evaluated is less than the fixed length, supplementing the password with a specific filling character or a coding value; if the length of the password to be evaluated exceeds the fixed length, intercepting the characters with the fixed length in front of the password to be evaluated;
encoding each character position of the password to be evaluated using a position encoding formula, the position encoding formula being as follows:
Wherein the method comprises the steps of Represent the firstThe position-coded value of the individual character,Represent the firstThe ASCII value of the individual character(s),Is the firstSine values of the individual character positions;
Combining the calculated position code values into a vector representation of a fixed length;
the fixed length vector representation of the password to be evaluated is normalized.
In a specific embodiment, the inputting the fixed-length vector representation of the password to be evaluated into the commercial password evaluation model, and the commercial password evaluation model outputting the evaluation result of the password to be evaluated further includes:
Screening commercial passwords which are most similar to the passwords to be evaluated from the vulnerable password set and the non-vulnerable password set based on the vector similarity and the password strength score;
And obtaining the evaluation result of the most similar commercial passwords and verifying the evaluation result of the password to be evaluated output by the model.
In a specific embodiment, the screening the commercial passwords from the vulnerable password set and the non-vulnerable password set that are most similar to the password to be evaluated based on the vector similarity and the password strength score includes:
Calculating vector similarity between the vector representation of the password to be evaluated and the vector representations of all commercial passwords in the vulnerable password set and the non-vulnerable password set using cosine similarity, and recording the vector similarity as first similarity ：
Wherein,Is a vector representation of the sample to be evaluated,Is a vector representation of a commercial password in the vulnerable password set and the non-vulnerable password set,AndRespectively representAndIs a norm of (2);
calculating the absolute value difference between the password intensity score of the password to be evaluated and the password intensity scores of all commercial passwords in the vulnerable password set and the non-vulnerable password set by using the following formula to be recorded as a second similarity ：
Wherein the method comprises the steps ofFor the password strength score of the password to be evaluated,The password strength score of a commercial password in the vulnerable password set and the non-vulnerable password set;
The final similarity between the password to be evaluated and all commercial passwords in the vulnerable password set and the non-vulnerable password set is calculated by using the following formula ：
Wherein the method comprises the steps ofIs a weight parameter between 0 and 1;
And calculating the final similarity between all commercial passwords in the vulnerable password set and the non-vulnerable password set and the password to be evaluated, wherein the commercial password corresponding to the maximum final similarity is the commercial password most similar to the password to be evaluated.
In a specific embodiment, the obtaining the evaluation result of the most similar commercial passwords and verifying the evaluation result of the model output password to be evaluated includes:
When the evaluation result of the password to be evaluated is that the vulnerability does not exist, if the evaluation result of the most similar commercial password does not exist, the evaluation result output by the model is the final evaluation result; if the evaluation result of the most similar commercial passwords is that the loopholes exist, the vector representation of the passwords to be evaluated is input to the commercial password evaluation model again, if the vector representation is input again for a preset number of times, the evaluation result still output by the model is that the loopholes do not exist, and the last evaluation result output by the model is taken as the final evaluation result; if the loopholes exist in the evaluation results which are output by the model for the first time after the model is input again within the preset times, judging that the loopholes exist in the evaluation results of the passwords to be evaluated, and combining the loopholes output by the model with the loopholes in the evaluation results of the most similar commercial passwords to be used as a final evaluation result;
When the evaluation result of the password to be evaluated is that the loophole exists, judging that the evaluation result of the password to be evaluated exists if the evaluation result of the most similar commercial password exists, and combining the loophole output by the model with the loophole in the evaluation result of the most similar commercial password to be used as a final evaluation result; if the evaluation result of the most similar commercial passwords is that the loopholes do not exist, the vector representation of the password to be evaluated is input to the commercial password evaluation model again, if the evaluation result still output by the model after the preset times of re-input is that the loopholes exist, the evaluation result output by the model last time is used as a final evaluation result; if the estimated results output by the model twice in succession after being input again within the preset times are no loopholes, the estimated result output by the model last time is taken as the final estimated result.
In a second aspect, the application provides a commercial password security evaluation system based on deep learning, which adopts the following technical scheme:
A deep learning based commercial password security assessment system comprising:
The system comprises a sample set acquisition module, a commercial password sample set acquisition module and a password analysis module, wherein the commercial password sample set comprises a vulnerable password set, a non-vulnerable password set, password intensities corresponding to the vulnerable password set and the non-vulnerable password set and vulnerability types corresponding to the vulnerable password set;
the sample set vector generation module is used for carrying out consistency processing on the vulnerable cipher set and the non-vulnerable cipher set by using a cipher conversion method based on a hash function, and generating vector representations of fixed lengths of the vulnerable cipher set and the non-vulnerable cipher set;
The evaluation model construction module is used for training a preset neural network model by taking the vector representations of the vulnerable password set and the fixed length of the non-vulnerable password set as input data and taking the corresponding password strength, the existence of a vulnerability and the specific vulnerability type as output to obtain a commercial password evaluation model;
The password vector generation module is used for acquiring a password to be evaluated, carrying out consistency processing on the password to be evaluated by using a password position coding method, and generating a vector representation with a fixed length of the password to be evaluated;
And the evaluation result output module is used for inputting the vector representation with the fixed length of the password to be evaluated into the commercial password evaluation model, and the commercial password evaluation model outputs the evaluation result of the password to be evaluated.
In a third aspect, the present application provides an electronic device comprising a processor and a memory; the memory stores therein a program that is loaded and executed by the processor to implement a commercial cryptographic security evaluation method based on deep learning as described in the first aspect.
In a fourth aspect, the present application provides a computer-readable storage medium having stored therein a program for implementing a deep learning-based commercial cryptographic security assessment method according to the first aspect when executed by a processor.
In summary, the beneficial effects of the present application at least include:
1) By training by using the deep learning model and evaluating by combining multidimensional information such as password strength, vulnerability type and the like, the comprehensiveness and precision of evaluation are greatly improved. The traditional method often depends on expert evaluation, but the method can be trained based on a large number of practical commercial password samples, so that subjectivity and limitation of the traditional method are overcome, and an evaluation result is more objective and accurate.
2) The model based on deep learning has the capability of dynamic real-time training, and can periodically update a commercial password sample set and retrain the model so as to cope with changes and emerging security holes in the password security environment. The real-time performance and the adaptability enable the evaluation result to always reflect the current password security state, and the coping capacity and the persistence of the system are improved.
3) And the password is subjected to consistency processing by adopting methods such as a hash function, password position coding and the like, so that a vector representation with a fixed length is generated, and the safety of the original password is effectively protected. The processing method reduces the risk of directly exposing the original password, provides a unified input format for the deep learning model on the premise of ensuring the data security, and improves the overall safety and the reliability of the evaluation system.
By collecting commercial password sample sets, including vulnerable and non-vulnerable passwords, each password is provided with a password strength and a vulnerability type tag. This step solves the limitation that traditional assessment methods rely on professional assessors and time sensitivity, ensuring the diversity and integrity of the assessment data. And secondly, carrying out consistency processing on the commercial passwords by adopting a password conversion method based on a hash function, and generating vector representation with fixed length. The processing improves the security and consistency of the password data, reduces the risk of directly exposing the original password, and simultaneously provides a unified input format for subsequent model training. And training the fixed-length vector representations by using a preset neural network model, wherein the input is the vector representation of the password, and the output is the strength of the password, whether the vulnerability exists or not and the evaluation result of the specific vulnerability type. The method solves the problems of insufficient accuracy and low efficiency in the traditional evaluation method, and improves the evaluation accuracy and speed through a deep learning technology. And finally, carrying out consistency processing on the password to be evaluated by using a password position coding method, generating a vector representation with a fixed length, and inputting the vector representation into a commercial password evaluation model. The model outputs the security evaluation result of the password to be evaluated, so that the security challenge of the commercial password in the digital environment is effectively solved, and the comprehensiveness and instantaneity of the password security evaluation are improved.
The foregoing description is only an overview of the present application, and is intended to provide a better understanding of the present application, as it is embodied in the following description, with reference to the preferred embodiments of the present application and the accompanying drawings.
Drawings
Fig. 1 is a schematic flow chart of a commercial password security assessment method based on deep learning in an embodiment of the application.
FIG. 2 is a flow chart of a method for performing consistency processing on a vulnerable cipher set and a non-vulnerable cipher set by using a cryptographic transformation method based on a hash function in an embodiment of the application.
FIG. 3 is an exemplary diagram of consistency processing of a vulnerable cipher set and a non-vulnerable cipher set using a hash function based filler cipher conversion method in an embodiment of the present application.
FIG. 4 is a flow chart of a consistency process of a password to be evaluated using a password position encoding method according to an embodiment of the present application.
FIG. 5 is an exemplary diagram of a consistency process for a password to be evaluated using a password position encoding method in an embodiment of the application.
Fig. 6 is an overall flowchart of a commercial password security assessment method based on deep learning in an embodiment of the application.
Fig. 7 is a block diagram of a commercial password security evaluation system based on deep learning in an embodiment of the application.
FIG. 8 is a block diagram of an electronic device for deep learning based commercial cryptographic security assessment in an embodiment of the application.
Detailed Description
The following describes in further detail the embodiments of the present application with reference to the drawings and examples. The following examples are illustrative of the application and are not intended to limit the scope of the application.
Optionally, the method for evaluating commercial password security based on deep learning provided by the embodiments of the present application is used for illustration in an electronic device, where the electronic device is a terminal or a server, and the terminal may be a computer, a tablet computer, etc., and the embodiment does not limit the type of the electronic device.
Referring to fig. 1, a flow chart of a method for evaluating commercial password security based on deep learning according to an embodiment of the present application includes at least the following steps:
step S101, a commercial password sample set is obtained, wherein the commercial password sample set comprises a vulnerable password set, a non-vulnerable password set, password intensities corresponding to the vulnerable password set and the non-vulnerable password set, and vulnerability types corresponding to the vulnerable password set.
Step S102, performing consistency processing on the vulnerable cipher set and the non-vulnerable cipher set by using a cipher conversion method based on a hash function, and generating vector representations of fixed lengths of the vulnerable cipher set and the non-vulnerable cipher set.
Step S103, training a preset neural network model by taking a fixed-length vector representation of the vulnerable password set and the non-vulnerable password set as input data and taking corresponding password strength, existence of a vulnerability and specific vulnerability type as output to obtain a commercial password evaluation model.
Step S104, obtaining the password to be evaluated, and carrying out consistency processing on the password to be evaluated by using a password position coding method to generate a vector representation with a fixed length of the password to be evaluated.
Step 105, a fixed-length vector representation of the password to be evaluated is input to a commercial password evaluation model, and the commercial password evaluation model outputs an evaluation result of the password to be evaluated.
According to the method, a commercial password sample set is collected, wherein the commercial password sample set comprises passwords with and without holes, and each password is provided with password strength and a vulnerability type label. This step solves the limitation that traditional assessment methods rely on professional assessors and time sensitivity, ensuring the diversity and integrity of the assessment data. And secondly, carrying out consistency processing on the commercial passwords by adopting a password conversion method based on a hash function, and generating vector representation with fixed length. The processing improves the security and consistency of the password data, reduces the risk of directly exposing the original password, and simultaneously provides a unified input format for subsequent model training. And training the fixed-length vector representations by using a preset neural network model, wherein the input is the vector representation of the password, and the output is the strength of the password, whether the vulnerability exists or not and the evaluation result of the specific vulnerability type. The method solves the problems of insufficient accuracy and low efficiency in the traditional evaluation method, and improves the evaluation accuracy and speed through a deep learning technology. And finally, carrying out consistency processing on the password to be evaluated by using a password position coding method, generating a vector representation with a fixed length, and inputting the vector representation into a commercial password evaluation model. The model outputs the security evaluation result of the password to be evaluated, so that the security challenge of the commercial password in the digital environment is effectively solved, and the comprehensiveness and instantaneity of the password security evaluation are improved.
In step S101, the commercial password sample set includes a vulnerable password set, a non-vulnerable password set, a strength of passwords corresponding to the vulnerable password set and the non-vulnerable password set, and a type of vulnerability corresponding to the vulnerable password set. The vulnerable password set comprises a plurality of commercial passwords with known security vulnerabilities, and the corresponding password strength and the vulnerability type of all commercial passwords in the vulnerable password set are recorded. The vulnerability-free password set comprises a plurality of commercial passwords which are known to have no security vulnerability, and simultaneously the password intensities corresponding to all commercial passwords in the vulnerability-free password set are recorded. For example, a commercial password with a security hole has a password strength score of 15 points, belongs to a weak password, and has a security hole with a leaked password. A commercial password without security holes has a password strength score of 60 minutes, belongs to a medium password, and does not have security holes. In implementation, all commercial passwords in the vulnerable password set and the non-vulnerable password set are subjected to strength evaluation through an open source password strength evaluation tool, so that corresponding password strength scores and corresponding strengths thereof are obtained, for example, zxcvbn and other evaluation tools.
Optionally, the commercial password sample set may be obtained from a public commercial password database and a security report, or may be obtained from research materials of commercial passwords, and the application is not limited to the obtaining manner of the commercial password sample set.
In addition, preferably, a pre-built distributed storage database is adopted to store and manage a commercial password sample set, a vulnerable password set and a non-vulnerable password set are stored and managed separately, and an open source password strength assessment tool is integrated into the distributed storage database, so that automatic assessment operation of password strength is realized. In addition, the distributed storage database periodically acquires a new commercial password sample set through an automatic script, and performs updating operation of the commercial password sample set.
Referring to fig. 2, a flow chart of consistency processing of a vulnerable cipher set and a non-vulnerable cipher set by using a cryptographic transformation method based on a hash function in an embodiment of the present application is shown, where the method at least includes the following steps:
Step S1021, a fixed length is set.
Where the fixed length will be the length of the finally generated vector representation. Alternatively, the fixed length is typically 16 characters, but may be selected to be 32 characters, and the application is not limited to the fixed length.
Step S1022, converting the commercial password into a hash value through a hash function.
In an implementation, all commercial passwords in the vulnerable and non-vulnerable password sets are hashed, and all commercial passwords are converted into hash values using a SHA-256 hash function, which is shown below:
wherein, The commercial code is represented by a code of commerce,The converted hash value is represented, typically 64 characters.
Alternatively, SHA-256 is selected as the hash function, and other types of hash functions, such as SHA-1 or MD5, may be selected, which is not limited by the specific type of hash function.
Step S1023, performing a transformation operation on the hash value by using the following transformation formula:
wherein, Representing a bitwise exclusive or operation,Representing the presentation to beThe two positions are shifted to the right,Representation fetchIs inverted by bit. Bitwise exclusive or operation is a commonly used bit operation that can introduce some randomness and confusion while maintaining data structure integrity. This helps to enhance the security of the cryptographic data so that the content of the original data cannot be easily inferred by simple observation. Shifting the hash value two bits to the right may cause a slight change in the information on each bit. This operation may increase the complexity of the hash value such that the transformed value is not easily reverse-extrapolated to the characteristics of the original data. The bit-wise inverting operation may change each bit of the hash value from 0 to 1 or from 1 to 0, which introduces a clear confusion effect, making it more difficult to directly link the transformed hash value with the original data. By these operations, the resultantNot only a simple hash value, but also a result after complex transformation, thus increasing the security of the cryptographic data, so that even if the transformed value is leaked, it is difficult for an attacker to infer the original cryptographic information. In addition, the transformation operation introduces nonlinearity and randomness, so that the expression capability of the data is improved, which is important for a machine learning model, because the input data can be more diversified and complicated, and the generalization capability of the model and the accuracy of safety evaluation are improved. The design of the transformation formula takes into account the requirements of security enhancement, data expression capability enhancement and input consistency. By such a design, hash values can be efficiently processed and vector representations with uniform length and high security can be generated, which is more suitable for cryptographic security assessment and other related applications.
Step S1024, the hash value after the transformation operation is intercepted into a vector representation with a fixed length.
Note that since the fixed length is set to 16 characters and the hash value is set to 64 characters in the present application, an operation of padding the vector of the hash value after the transformation operation does not occur.
Step S1025, the normalization process is performed on the vector representation.
In implementation, by performing normalization processing on the truncated vector representations, the value ranges of all vector representations can be made to be within a certain range as much as possible. The normalization process helps to prevent the influence of too large or too small a numerical range on subsequent model training, and enhances the stability and consistency of data.
Referring to fig. 3, an exemplary diagram of performing consistency processing on a vulnerable cipher set and a non-vulnerable cipher set by using a hash function-based filler cipher conversion method in an embodiment of the present application is shown, where processing procedures of the vulnerable cipher and the non-vulnerable cipher are respectively shown. In step S102, since training of the model is involved in the following, all commercial passwords in the flawed password set and the flawed password set are converted into vector representations of fixed length, so that input data of the following model have the same dimension and length, and data of uniform length not only makes data processing of each batch more efficient, because dynamic adjustment in the batch is not required. And the fixed-length data can fully utilize vectorization and parallel computing technologies, so that the training and reasoning speed is improved, and the waste of computing resources is reduced.
In step S103, the commercial password evaluation model generated by training the preset neural network model with the commercial password sample set is dynamically trained in real time, and since the commercial password sample set is updated periodically, the commercial password evaluation model is trained by re-acquiring and new commercial password sample set at intervals, so as to ensure the training strength and evaluation accuracy of the model as much as possible.
Optionally, the convolutional neural network is selected as a preset neural network model, and because the convolutional neural network has better performance in a natural language processing task, other types of neural networks can be used, and the application does not limit the specific types of the neural network model.
In step S104, the cryptographic position coding method is a method of converting a password into a fixed-length vector representation by character-by-character processing, position coding, flexible padding, and on-the-fly normalization. Referring to fig. 4, a flow chart of a consistency process of a password to be evaluated by using a password position encoding method according to an embodiment of the present application is shown, where the method at least includes the following steps:
s1041, if the length of the password to be evaluated is less than the fixed length, supplementing it with a specific pad character or code value. And if the length of the password to be evaluated exceeds the fixed length, intercepting the characters with the fixed length in front of the password to be evaluated.
In practice, if the length of the password to be evaluated is less than the fixed length set, it is necessary to supplement it with specific pad characters or code values to ensure that the fixed length requirement is met. The pad character may be selected as any character, typically a space or other symbol. The choice of pad characters should not affect the subsequent position-coding process because the position-coding formula depends mainly on the ASCII value and position information of the characters. If the length of the password to be evaluated exceeds the set fixed length, the character part of the former fixed length of the password to be evaluated needs to be intercepted. This ensures to some extent that the subsequent position encoding and vector representation generation steps can handle the same length of data, avoiding unnecessary complexity and data processing problems.
S1042, coding each character position of the password to be evaluated by using a position coding formula.
Specifically, the position coding formula is as follows:
Wherein the method comprises the steps of Represent the firstThe position-coded value of the individual character,Represent the firstThe ASCII value of the individual character(s),Is the firstSine values of the character positions, and sine functions in the design process of the position coding formulaThe introduction of non-linear features helps introduce some variation and randomness in the position coding, increasing the diversity of the vector representation.ASCII values representing characters directly express basic information of the charactersCoding information representing the character position can help the model understand the relative position of the character in the password.
S1043, combining the calculated position code values into a vector representation with a fixed length.
Specifically, the position code value of each character is calculatedA vector representation of fixed length is composed.
S1044, carrying out standardization processing on the vector representation of the fixed length of the password to be evaluated.
Specifically, the generated fixed-length vector representations are normalized to ensure that all vector values are within the same range as much as possible.
In step S104, referring to fig. 5, an exemplary diagram of performing a consistency process on a password to be evaluated by using a password position encoding method in an embodiment of the present application is shown, where the process of the password to be evaluated is shown respectively.
Furthermore, preferably, the processing method in the step S104 is applicable to single data, each password is processed independently, the position coding formula generates a unique position coding value for each character, and the design retains specific position information of each character in the password and is not affected by other passwords. This is critical to the processing of individual cryptographic data, as it ensures that each cipher has its unique characteristic representation, not confused with the processing of other ciphers. The processing method of step S102 is applicable to a plurality of data, and the hash function can be used to quickly convert a large number of commercial passwords into hash values with fixed lengths. The calculation of the hash function is generally efficient and can complete the conversion operation in a reasonable time even if a large amount of data is processed. The transformation formula introduces operations such as bitwise exclusive or, right shift, bitwise inversion and the like, and increases the complexity and randomness of the hash value. This complexity makes it difficult for the generated vector representation to be inversely extrapolated to the original cryptographic information, even when large amounts of data are processed, thereby enhancing the security of the data.
In step S105, the vector representation of the password to be evaluated is input to a commercial password evaluation model, the commercial password evaluation model outputs an evaluation result of the password to be evaluated, the evaluation result includes a password strength score of the password to be evaluated and a corresponding strength thereof, if no vulnerability exists, no vulnerability is output, and if a vulnerability exists, a specific vulnerability type is output.
Referring to fig. 6, an overall flow diagram of a commercial password security assessment method based on deep learning according to an embodiment of the present application is provided. In this embodiment, the commercial password security evaluation method includes, in addition to the foregoing steps S101 to S105, the steps of:
and S106, screening commercial passwords which are most similar to the passwords to be evaluated from the vulnerable password set and the non-vulnerable password set based on the vector similarity and the password intensity score.
Specifically, first, the cosine similarity is used to calculate the vector similarity between the vector representation of the password to be evaluated and the vector representations of all commercial passwords in the vulnerable password set and the non-vulnerable password set, and the vector similarity is recorded as the first similarity：
Wherein,Is a vector representation of the sample to be evaluated,Is a vector representation of a commercial password in the vulnerable password set and the non-vulnerable password set,AndRespectively representAndIs a norm of (c). It should be noted that here the vector representations of both are still at the same fixed length. The absolute value difference between the password intensity score of the password to be evaluated and the password intensity scores of all commercial passwords in the vulnerable password set and the non-vulnerable password set is calculated by using the following formula and recorded as a second similarity：
Wherein the method comprises the steps ofFor the password strength score of the password to be evaluated,The password strength score of a commercial password in the vulnerable password set and the non-vulnerable password set. Finally, the final similarity between the password to be evaluated and all commercial passwords in the vulnerable password set and the vulnerable password set is calculated by using the following formula：
Wherein the method comprises the steps ofIs a weight parameter between 0 and 1 for balancing the effects of cosine similarity and password strength score differences. The above formula takes into account both cosine similarity and password strength score differences. Cosine similarity reflects geometric similarity between vector representations, while password strength score differences reflect differences in security and complexity of passwords. Such comprehensive consideration may more fully evaluate the similarity between passwords. And calculating the final similarity between all commercial passwords in the vulnerable password set and the non-vulnerable password set and the password to be evaluated in the calculation mode, wherein the commercial password corresponding to the maximum final similarity is the commercial password most similar to the password to be evaluated. The commercial passwords which are most similar are judged only through the similarity of the password structures, so that the password strength score is introduced, the password strength of the commercial passwords which are most similar is matched with that of the password to be evaluated as much as possible, and the commercial passwords which are most similar are found out are more accurate to a certain extent.
Further, the present application preferably calculates the first similarityIn addition to the above method, the following methods may be used to obtain better effects: firstly, calculating the vector similarity between the vector representation of the password to be evaluated and the vector representations of all commercial passwords in the vulnerable password set and the non-vulnerable password set by using cosine similarity, and then calculating the vector similarity between the vector representation of the password to be evaluated and the vector representations of all commercial passwords in the vulnerable password set and the non-vulnerable password set by using a Gaussian kernel function, wherein the calculation formula is as follows:
wherein, Is a parameter controlling the decay rate of the kernel,Is a vector representation of the sample to be evaluated,Is a vector representation of a commercial password in both the vulnerable and non-vulnerable password sets. Finally, the sum of the two calculated vector similarity sums is averaged to be used as the first similarity. The cosine similarity and the kernel function are combined to obtain the value of the first similarity, and the cosine similarity measures the geometric similarity between vectors and is suitable for vectors in Euclidean space. While kernel functions may perform nonlinear mapping to map data into higher dimensional space to capture more complex relationships. In combination, both geometric features and higher dimensional features can be considered in evaluating similarity. Both cosine similarity and kernel functions can be mapped into a range of values from 0 to 1, which allows their results to be intuitively interpreted as a measure of similarity, with larger values representing higher similarity. The larger and more similar the cosine similarity and the nature of the kernel function, i.e. the higher the similarity is indicated when the value is close to 1, which is consistent with the intuitive understanding of the similarity measure in practical applications. The cosine similarity or kernel function alone may exhibit limitations due to differences in data characteristics. The combination of the two can make up the respective defects and improve the comprehensiveness and accuracy of similarity evaluation. In summary, by using cosine similarity calculation first, then mapping by using a kernel function, and finally averaging, geometric features and features with higher dimensionality can be effectively combined, and the comprehensiveness and accuracy of similarity evaluation can be improved.
Step S107, obtaining the evaluation result of the most similar commercial passwords and verifying the evaluation result of the password to be evaluated output by the model.
Specifically, when the evaluation result of the password to be evaluated output by the model is that no vulnerability exists, if the evaluation result of the most similar commercial password is that no vulnerability exists, the evaluation result output by the model is the final evaluation result. If the evaluation result of the most similar commercial passwords is that the loopholes exist, the vector representation of the password to be evaluated is input to the commercial password evaluation model again, if the evaluation result still output by the model after the preset times of re-input is that the loopholes do not exist, the last evaluation result output by the model is taken as the final evaluation result. If the loopholes exist in the evaluation results which are output by the model for the first time after the loopholes are input again in the preset times, judging that the loopholes exist in the evaluation results of the passwords to be evaluated, and combining the loopholes output by the model with the loopholes in the evaluation results of the most similar commercial passwords to be used as the final evaluation results.
When the evaluation result of the password to be evaluated is the existence of the loophole, if the evaluation result of the most similar commercial password is the existence of the loophole, judging that the evaluation result of the password to be evaluated is the existence of the loophole, and combining the loophole output by the model with the loophole in the evaluation result of the most similar commercial password to be used as a final evaluation result. If the evaluation result of the most similar commercial passwords is that the loopholes do not exist, the vector representation of the password to be evaluated is input to the commercial password evaluation model again, if the evaluation result still output by the model after the preset times of re-input is that the loopholes exist, the last evaluation result output by the model is taken as the final evaluation result. If the estimated results output by the model twice in succession after being input again within the preset times are no loopholes, the estimated result output by the model last time is taken as the final estimated result.
Fig. 7 is a block diagram of a commercial password security evaluation system based on deep learning according to an embodiment of the present application. The device at least comprises the following modules:
The sample set acquisition module is used for acquiring a commercial password sample set, wherein the commercial password sample set comprises a vulnerable password set, a non-vulnerable password set, the password strength corresponding to the vulnerable password set and the non-vulnerable password set, and the vulnerability type corresponding to the vulnerable password set.
The sample set vector generation module is used for carrying out consistency processing on the vulnerable cipher set and the non-vulnerable cipher set by using a cipher conversion method based on a hash function, and generating vector representations with fixed lengths of the vulnerable cipher set and the non-vulnerable cipher set.
The evaluation model construction module is used for training a preset neural network model by taking a fixed-length vector representation of the leaky password set and the non-leaky password set as input data and taking corresponding password strength, existence of a leak and specific leak type as output to obtain a commercial password evaluation model.
The password vector generation module is used for acquiring the password to be evaluated, carrying out consistency processing on the password to be evaluated by using a password position coding method, and generating a vector representation with fixed length of the password to be evaluated.
The evaluation result output module is used for inputting the vector representation of the fixed length of the password to be evaluated into the commercial password evaluation model, and the commercial password evaluation model outputs the evaluation result of the password to be evaluated.
For relevant details reference is made to the method embodiments described above.
Fig. 8 is a block diagram of an electronic device provided in one embodiment of the application. The device comprises at least a processor 401 and a memory 402.
Processor 401 may include one or more processing cores such as: 4 core processors, 8 core processors, etc. The processor 401 may be implemented in at least one hardware form of DSP (DIGITAL SIGNAL Processing), FPGA (Field-Programmable gate array), PLA (Programmable Logic Array ). Processor 401 may also include a main processor, which is a processor for processing data in an awake state, also called a CPU (Central Processing Unit ), and a coprocessor; a coprocessor is a low-power processor for processing data in a standby state. In some embodiments, the processor 401 may integrate a GPU (Graphics Processing Unit, image processor) for rendering and drawing of content required to be displayed by the display screen. In some embodiments, the processor 401 may also include an AI (ARTIFICIAL INTELLIGENCE ) processor for processing computing operations related to machine learning.
Memory 402 may include one or more computer-readable storage media, which may be non-transitory. Memory 402 may also include high-speed random access memory, as well as non-volatile memory, such as one or more magnetic disk storage devices, flash memory storage devices. In some embodiments, a non-transitory computer readable storage medium in memory 402 is used to store at least one instruction for execution by processor 401 to implement the deep learning based commercial cryptographic security assessment method provided by the method embodiments of the present application.
In some embodiments, the electronic device may further optionally include: a peripheral interface and at least one peripheral. The processor 401, memory 402, and peripheral interfaces may be connected by buses or signal lines. The individual peripheral devices may be connected to the peripheral device interface via buses, signal lines or circuit boards. Illustratively, peripheral devices include, but are not limited to: radio frequency circuitry, touch display screens, audio circuitry, and power supplies, among others.
Of course, the electronic device may also include fewer or more components, as the present embodiment is not limited in this regard.
Optionally, the present application further provides a computer readable storage medium, in which a program is stored, and the program is loaded and executed by a processor to implement the method for evaluating commercial password security based on deep learning according to the above method embodiment.
Optionally, the present application further provides a computer product, where the computer product includes a computer readable storage medium, where a program is stored, and the program is loaded and executed by a processor to implement the method for evaluating commercial cryptographic security based on deep learning according to the above method embodiment.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (10)
1. A method for evaluating security of a commercial password based on deep learning, the method comprising:
Obtaining a commercial password sample set, wherein the commercial password sample set comprises a vulnerable password set, a non-vulnerable password set, password intensities corresponding to the vulnerable password set and the non-vulnerable password set and vulnerability types corresponding to the vulnerable password set;
Performing consistency processing on the vulnerable cipher set and the non-vulnerable cipher set by using a cipher conversion method based on a hash function, and generating vector representations of fixed lengths of the vulnerable cipher set and the non-vulnerable cipher set;
Training a preset neural network model by taking the vector representations of the fixed lengths of the vulnerable cipher set and the non-vulnerable cipher set as input data and taking the corresponding cipher strength, the existence of the vulnerability and the specific vulnerability type as output to obtain a commercial cipher evaluation model;
Obtaining a password to be evaluated, performing consistency processing on the password to be evaluated by using a password position coding method, and generating a vector representation of a fixed length of the password to be evaluated;
And inputting the vector representation with the fixed length of the password to be evaluated into the commercial password evaluation model, and outputting an evaluation result of the password to be evaluated by the commercial password evaluation model.
2. The deep learning based commercial cryptographic security assessment method of claim 1, wherein the performing a consistency process on the vulnerable cryptographic set and the non-vulnerable cryptographic set using a cryptographic transformation method based on a hash function, generating a vector representation of fixed lengths of the vulnerable cryptographic set and the non-vulnerable cryptographic set comprises:
Setting a fixed length;
Converting the commercial password into a hash value through a hash function;
performing transformation operation on the hash value by using a transformation formula;
Intercepting the hash value after the transformation operation into a vector representation with a fixed length;
The vector representation is normalized.
3. The deep learning-based commercial password security assessment method according to claim 2, wherein the converting the commercial password into a hash value by a hash function, and performing a transformation operation on the hash value using a transformation formula comprises:
all commercial passwords are converted to hash values using a SHA-256 hash function, which is shown below:
wherein, The commercial code is represented by a code of commerce,Representing the converted hash value;
Transforming the hash value using a transformation formula:
wherein, Representing a bitwise exclusive or operation,Representing the presentation to beThe two positions are shifted to the right,Representation fetchIs inverted by bit.
4. The deep learning based commercial password security assessment method of claim 1, wherein the obtaining a password to be assessed, performing consistency processing on the password to be assessed using a password position coding method, generating a fixed-length vector representation of the password to be assessed comprises:
If the length of the password to be evaluated is less than the fixed length, supplementing the password with a specific filling character or a coding value; if the length of the password to be evaluated exceeds the fixed length, intercepting the characters with the fixed length in front of the password to be evaluated;
encoding each character position of the password to be evaluated using a position encoding formula, the position encoding formula being as follows:
Wherein the method comprises the steps of Represent the firstThe position-coded value of the individual character,Represent the firstThe ASCII value of the individual character(s),Is the firstSine values of the individual character positions;
Combining the calculated position code values into a vector representation of a fixed length;
the fixed length vector representation of the password to be evaluated is normalized.
5. The deep learning-based commercial password security assessment method according to claim 1, wherein the inputting the fixed-length vector representation of the password to be assessed into the commercial password assessment model, after outputting the assessment result of the password to be assessed, further comprises:
Screening commercial passwords which are most similar to the passwords to be evaluated from the vulnerable password set and the non-vulnerable password set based on the vector similarity and the password strength score;
And obtaining the evaluation result of the most similar commercial passwords and verifying the evaluation result of the password to be evaluated output by the model.
6. The deep learning based commercial password security assessment method according to claim 5, wherein the screening commercial passwords from the vulnerable password set and the non-vulnerable password set based on the vector similarity and the password strength score comprises:
Calculating vector similarity between the vector representation of the password to be evaluated and the vector representations of all commercial passwords in the vulnerable password set and the non-vulnerable password set using cosine similarity, and recording the vector similarity as first similarity ：
Wherein,Is a vector representation of the sample to be evaluated,Is a vector representation of a commercial password in the vulnerable password set and the non-vulnerable password set,AndRespectively representAndIs a norm of (2);
calculating the absolute value difference between the password intensity score of the password to be evaluated and the password intensity scores of all commercial passwords in the vulnerable password set and the non-vulnerable password set by using the following formula to be recorded as a second similarity ：
Wherein the method comprises the steps ofFor the password strength score of the password to be evaluated,The password strength score of a commercial password in the vulnerable password set and the non-vulnerable password set;
The final similarity between the password to be evaluated and all commercial passwords in the vulnerable password set and the non-vulnerable password set is calculated by using the following formula ：
Wherein the method comprises the steps ofIs a weight parameter between 0 and 1;
And calculating the final similarity between all commercial passwords in the vulnerable password set and the non-vulnerable password set and the password to be evaluated, wherein the commercial password corresponding to the maximum final similarity is the commercial password most similar to the password to be evaluated.
7. The deep learning based commodity password security assessment method according to claim 5, wherein the obtaining the assessment result of the most similar commodity password and verifying the assessment result of the model output password to be assessed comprises:
When the evaluation result of the password to be evaluated is that the vulnerability does not exist, if the evaluation result of the most similar commercial password does not exist, the evaluation result output by the model is the final evaluation result; if the evaluation result of the most similar commercial passwords is that the loopholes exist, the vector representation of the passwords to be evaluated is input to the commercial password evaluation model again, if the vector representation is input again for a preset number of times, the evaluation result still output by the model is that the loopholes do not exist, and the last evaluation result output by the model is taken as the final evaluation result; if the loopholes exist in the evaluation results which are output by the model for the first time after the model is input again within the preset times, judging that the loopholes exist in the evaluation results of the passwords to be evaluated, and combining the loopholes output by the model with the loopholes in the evaluation results of the most similar commercial passwords to be used as a final evaluation result;
When the evaluation result of the password to be evaluated is that the loophole exists, judging that the evaluation result of the password to be evaluated exists if the evaluation result of the most similar commercial password exists, and combining the loophole output by the model with the loophole in the evaluation result of the most similar commercial password to be used as a final evaluation result; if the evaluation result of the most similar commercial passwords is that the loopholes do not exist, the vector representation of the password to be evaluated is input to the commercial password evaluation model again, if the evaluation result still output by the model after the preset times of re-input is that the loopholes exist, the evaluation result output by the model last time is used as a final evaluation result; if the estimated results output by the model twice in succession after being input again within the preset times are no loopholes, the estimated result output by the model last time is taken as the final estimated result.
8. A deep learning-based commercial password security assessment system, comprising:
The system comprises a sample set acquisition module, a commercial password sample set acquisition module and a password analysis module, wherein the commercial password sample set comprises a vulnerable password set, a non-vulnerable password set, password intensities corresponding to the vulnerable password set and the non-vulnerable password set and vulnerability types corresponding to the vulnerable password set;
the sample set vector generation module is used for carrying out consistency processing on the vulnerable cipher set and the non-vulnerable cipher set by using a cipher conversion method based on a hash function, and generating vector representations of fixed lengths of the vulnerable cipher set and the non-vulnerable cipher set;
The evaluation model construction module is used for training a preset neural network model by taking the vector representations of the vulnerable password set and the fixed length of the non-vulnerable password set as input data and taking the corresponding password strength, the existence of a vulnerability and the specific vulnerability type as output to obtain a commercial password evaluation model;
The password vector generation module is used for acquiring a password to be evaluated, carrying out consistency processing on the password to be evaluated by using a password position coding method, and generating a vector representation with a fixed length of the password to be evaluated;
And the evaluation result output module is used for inputting the vector representation with the fixed length of the password to be evaluated into the commercial password evaluation model, and the commercial password evaluation model outputs the evaluation result of the password to be evaluated.
9. An electronic device comprising a processor and a memory; the memory stores therein a program that is loaded and executed by the processor to implement a deep learning-based commercial cryptographic security assessment method as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium, in which a program is stored which, when executed by a processor, is adapted to carry out a deep learning-based commercial cryptographic security assessment method as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411027828.6A CN118551364B (en) | 2024-07-30 | 2024-07-30 | Commercial password security assessment method and system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411027828.6A CN118551364B (en) | 2024-07-30 | 2024-07-30 | Commercial password security assessment method and system based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118551364A true CN118551364A (en) | 2024-08-27 |
| CN118551364B CN118551364B (en) | 2024-10-01 |
Family
ID=92453487
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411027828.6A Active CN118551364B (en) | 2024-07-30 | 2024-07-30 | Commercial password security assessment method and system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118551364B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110362987A (en) * | 2019-06-29 | 2019-10-22 | 南京理工大学 | A kind of lightweight assessment algorithm of Cipher Strength |
| CN113010895A (en) * | 2020-12-08 | 2021-06-22 | 四川大学 | Vulnerability hazard assessment index technology based on deep learning |
| CN113918471A (en) * | 2021-11-19 | 2022-01-11 | 中国工商银行股份有限公司 | Test case processing method and device and computer readable storage medium |
| CN114021450A (en) * | 2021-11-02 | 2022-02-08 | 湖州师范学院 | Method for predicting reliability of circuit of double-self-encoder by fusing Hash coding and blocking strategies |
| CN115422527A (en) * | 2022-07-20 | 2022-12-02 | 武汉船舶通信研究所（中国船舶重工集团公司第七二二研究所） | Password strength evaluation method and device and server |
| CN117319091A (en) * | 2023-11-29 | 2023-12-29 | 苏州市软件评测中心有限公司 | Enterprise software network security vulnerability detection method and system based on deep learning |
| CN118364479A (en) * | 2023-10-24 | 2024-07-19 | 朱彪 | Method for encrypting password by using trigonometric function |
-
2024
- 2024-07-30 CN CN202411027828.6A patent/CN118551364B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110362987A (en) * | 2019-06-29 | 2019-10-22 | 南京理工大学 | A kind of lightweight assessment algorithm of Cipher Strength |
| CN113010895A (en) * | 2020-12-08 | 2021-06-22 | 四川大学 | Vulnerability hazard assessment index technology based on deep learning |
| CN114021450A (en) * | 2021-11-02 | 2022-02-08 | 湖州师范学院 | Method for predicting reliability of circuit of double-self-encoder by fusing Hash coding and blocking strategies |
| CN113918471A (en) * | 2021-11-19 | 2022-01-11 | 中国工商银行股份有限公司 | Test case processing method and device and computer readable storage medium |
| CN115422527A (en) * | 2022-07-20 | 2022-12-02 | 武汉船舶通信研究所（中国船舶重工集团公司第七二二研究所） | Password strength evaluation method and device and server |
| CN118364479A (en) * | 2023-10-24 | 2024-07-19 | 朱彪 | Method for encrypting password by using trigonometric function |
| CN117319091A (en) * | 2023-11-29 | 2023-12-29 | 苏州市软件评测中心有限公司 | Enterprise software network security vulnerability detection method and system based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118551364B (en) | 2024-10-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kong et al. | Memristor-induced hyperchaos, multiscroll and extreme multistability in fractional-order HNN: Image encryption and FPGA implementation | |
| Trujillo-Toledo et al. | Real-time RGB image encryption for IoT applications using enhanced sequences from chaotic maps | |
| Chen et al. | Pseudorandom Number Generator Based on Three Kinds of Four‐Wing Memristive Hyperchaotic System and Its Application in Image Encryption | |
| US9787647B2 (en) | Secure computer evaluation of decision trees | |
| CN110245469B (en) | Webpage watermark generation method, watermark analysis method, device and storage medium | |
| TW202107305A (en) | Model parameter determination method and device and electronic equipment | |
| Song et al. | Privacy-preserving threshold-based image retrieval in cloud-assisted Internet of Things | |
| CN108400868A (en) | Storage method, device and the mobile terminal of seed key | |
| US20160133341A1 (en) | Signal transition analysis of a circuit | |
| Liu et al. | Target-based image encryption via infinite interval chaotic system with Ill-conditioned parameter and 3DBDM | |
| CN112149141A (en) | Model training method, device, equipment and medium | |
| CN110069907A (en) | Big data source tracing method and system based on digital watermarking | |
| Liu et al. | A novel security key generation method for SRAM PUF based on Fourier analysis | |
| Wang et al. | A privacy-preserving cross-media retrieval on encrypted data in cloud computing | |
| CN118551364B (en) | Commercial password security assessment method and system based on deep learning | |
| US20240113888A1 (en) | Post-quantum lattice-based signature latency reduction | |
| Demmese et al. | Machine learning based fileless malware traffic classification using image visualization | |
| AlKhafaji et al. | Improved technique for hiding data in a colored and a monochrome image | |
| CN112149834A (en) | Model training method, device, equipment and medium | |
| Agarwal et al. | Analysis of Bao-Zhou-Chen-Liu’s hybrid chaotic system | |
| Pal et al. | Efficient image encryption using the Tinkerbell map in conjunction with linear feedback shift registers | |
| Baladhay et al. | AES-128 reduced-round permutation by replacing the MixColumns function | |
| Chen et al. | Detection Method of Hardware Trojan Based on Attention Mechanism and Residual-Dense-Block under the Markov Transition Field | |
| Su et al. | Chaotic image encryption algorithm based on bit-level feedback adjustment | |
| Huang et al. | Performance of finite precision on discrete Chaotic map based on a feedback shift register |
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 |