Intelligent screening method for software information of motor vehicle inspection and testing equipment

Abstract

The present application relates to the field of intelligent data analysis, and provides a method for intelligent screening of software information of motor vehicle inspection and testing equipment, which extracts the source code in the workstation control exe software package to obtain the workstation control software source code, and uses source code processing and analysis technology based on deep learning to perform sentence processing, semantic coding and optimization of the workstation control software source code, so as to automatically obtain security vulnerability detection results based on the global correlation information of the semantic features of each optimized source code. In this way, potential vulnerabilities in the source code can be more accurately identified, including those problems that will only appear under specific runtime conditions. At the same time, it can understand and analyze complex business logic, so as to discover logical errors hidden in deep conditional statements and program flows, thereby improving the intelligence level of software vulnerability detection.

Description

Technical Field

[0001] The present application relates to the field of intelligent data analysis, and more specifically, to a method for intelligently screening software information of motor vehicle inspection and testing equipment. Background Art

[0002] Motor vehicle inspection and testing equipment is a professional device used to conduct various safety and performance inspections on motor vehicles. These devices usually include hardware and software parts, of which the software part is responsible for controlling the operation of the hardware, data collection, processing, and report generation.

[0003] The control software of the workstation is one of the core components of this type of equipment, which is directly related to the accuracy, reliability and safety of the test. The control software of the workstation is responsible for coordinating the test process, managing the test data, and ensuring the accuracy and reliability of the test results. Therefore, the security and stability of the software are of vital importance. Any software loopholes may lead to incorrect test results, which in turn affects the safety assessment of the motor vehicle and even leads to legal disputes.

[0004] However, traditional static code analysis tools focus mainly on the syntax and structure of the code, rather than the actual execution path of the code, in the security assessment of workstation control software. This means that static analysis may not provide sufficient insights for vulnerabilities that only manifest under specific runtime conditions, such as time-related race conditions, memory leaks, or complex input processing logic. In addition, static analysis tools often have difficulty understanding and evaluating complex business logic, which may cause them to miss logical errors hidden in deep conditional statements and program flows. For example, a complex conditional branch may lead to unauthorized access or data leakage under certain conditions, but these conditions may not be fully explored during static analysis.

[0005] Therefore, an optimized method for screening software information of motor vehicle inspection and testing equipment is desired. Summary of the invention

[0006] In view of the shortcomings of the prior art, the present application provides a method for intelligently screening software information of motor vehicle inspection and testing equipment.

[0007] According to one aspect of the present application, a method for intelligently screening software information of motor vehicle inspection and testing equipment is provided, which includes:

[0008] Determine the directory location and file location of the motor vehicle inspection and testing equipment software;

[0009] Based on the directory location and the file location, extract the workstation control exe software package;

[0010] Extract key information from the workstation control exe software package, the key information including version number, compilation date, digital signature and configuration file;

[0011] Perform vulnerability detection on the workstation control exe software package to obtain a security vulnerability detection result;

[0012] Among them, performing vulnerability detection on the workstation control exe software package to obtain security vulnerability detection results includes:

[0013] Extract the source code in the workstation control exe software package to obtain the workstation control software source code;

[0014] Sentence processing is performed on the workstation control software source code to obtain a sequence of description sentences of the workstation control software source code;

[0015] Semantically encoding each workstation control software source code description sentence in the sequence of workstation control software source code description sentences to obtain a sequence of granular semantic encoding feature vectors of the workstation control software source code sentences;

[0016] Performing feature association optimization processing based on sequence endogenous correlation quantization on the sequence of granular semantic coding feature vectors of the workstation control software source code sentence to obtain a sequence of granular semantic optimized coding feature vectors of the workstation control software source code sentence;

[0017] Inputting the sequence of the station machine control software source code sentence granularity semantic optimization coding feature vectors into the code sentence granularity global semantic association encoder to obtain the station machine control software source code global semantic coding feature vector;

[0018] The security vulnerability detection result is obtained based on the global semantic encoding feature vector of the workstation control software source code.

[0019] This application has significant technical effects due to the adoption of the above technical solutions:

[0020] The present application provides an intelligent screening method for software information of motor vehicle inspection and testing equipment, which extracts the source code in the workstation control exe software package to obtain the workstation control software source code, and uses source code processing and analysis technology based on deep learning to perform sentence processing, semantic coding and optimization of the workstation control software source code, so as to automatically obtain security vulnerability detection results based on the global correlation information of the semantic features of each optimized source code. In this way, potential vulnerabilities in the source code can be more accurately identified, including those problems that will only appear under specific runtime conditions. At the same time, it is possible to understand and analyze complex business logic, thereby discovering logical errors hidden in deep conditional statements and program flows, thereby improving the intelligence level of software vulnerability detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0022] Figure 1 The present invention is a flowchart of a method for intelligently screening software information of motor vehicle inspection and testing equipment according to an embodiment of the present application.

[0023] Figure 2 The present invention is a flowchart of performing vulnerability detection on the workstation control exe software package to obtain security vulnerability detection results in the method for intelligently screening software information of motor vehicle inspection and testing equipment according to an embodiment of the present application.

[0024] Figure 3 This is a data flow diagram for performing vulnerability detection on the workstation machine control exe software package to obtain security vulnerability detection results in the method for intelligent screening of software information of motor vehicle inspection and testing equipment according to an embodiment of the present application.

[0025] Figure 4 In the method for intelligent screening of software information of motor vehicle inspection and testing equipment according to an embodiment of the present application, a flowchart is provided in which a sequence of granular semantic encoding feature vectors of the source code sentence of the workstation control software is input into a feature context semantic association optimization module based on quantification of the sequence's intrinsic correlation to obtain a sequence of granular semantic optimization encoding feature vectors of the source code sentence of the workstation control software. DETAILED DESCRIPTION

[0026] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0027] It should be understood that the various steps described in the method implementation of the present application can be performed in different orders and / or in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0028] In the description of the embodiments of the present application, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0029] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0030] Motor vehicle inspection and testing equipment is a precision tool designed specifically for safety and performance testing of vehicles. These devices consist of two major parts: hardware and software. The software part plays a vital role, not only controlling the operation of the hardware, but also responsible for data collection, processing and report generation.

[0031] In this type of equipment, the station control software is the core component, which directly determines the accuracy and reliability of the test. The main task of the software is to manage the entire test process, maintain the test data, and ensure the accuracy of the test results. The security and stability of the software are crucial to the test results. Any potential software defects may lead to erroneous test results, which not only affects the safety assessment of motor vehicles, but may also cause legal issues.

[0032] Although traditional static code analysis tools play a role in assessing the security of workstation control software, they mainly focus on the syntax and structure analysis of the code rather than the actual execution path of the code. This means that static analysis may not provide sufficient insight into issues that only appear under specific operating conditions, such as time-related race conditions, memory leaks, or complex input processing logic. In addition, static analysis tools have limitations in understanding and evaluating complex business logic, which may cause them to ignore logical defects hidden in deep conditional judgments and program flows. For example, a complex conditional branch may lead to unauthorized access or data leakage under certain conditions, but these conditions may not be fully considered during the static analysis process.

[0033] Based on this, the present application proposes a method for intelligently screening software information of motor vehicle inspection and testing equipment. Figure 1 Flow chart of the method for intelligently screening software information of motor vehicle inspection and testing equipment according to an embodiment of the present application. Figure 1As shown, according to the embodiment of the present application, the method for intelligent screening of software information of motor vehicle inspection and testing equipment includes: S110, determining the directory location and file location of the motor vehicle inspection and testing equipment software; S120, extracting the workstation control exe software package based on the directory location and the file location; S130, extracting key information from the workstation control exe software package, the key information including version number, compilation date, digital signature and configuration file; S140, performing vulnerability detection on the workstation control exe software package to obtain security vulnerability detection results.

[0034] It should be understood that in order to clarify the specific target range that needs to be analyzed and processed, it is first necessary to determine the directory location and file location of the motor vehicle inspection and detection equipment software. Generally, the specified installation path can be scanned through the API of the operating system to determine the software directory location and file location. Next, considering that the motor vehicle inspection and detection equipment software part is relatively complex, it contains multiple modules and files with different functions. The station machine control exe software package is the core component of the motor vehicle inspection and detection equipment software, which contains the key components and codes required for the software to run. In order to ensure that no key information is missed during the analysis process, it is necessary to extract the station machine control exe software package based on the directory location and the file location. Then, in order to comprehensively evaluate the security of the software, it is necessary to extract key information such as version number, compilation date, digital signature and configuration file from the station machine control exe software package. This information provides a basis for vulnerability management and troubleshooting. By systematically extracting and analyzing this information, the overall reliability and security of the motor vehicle inspection and detection equipment software can be effectively improved. Finally, it is necessary to perform vulnerability detection on the station machine control exe software package to obtain security vulnerability detection results.

[0035] Accordingly, in the process of performing vulnerability detection on the workstation control exe software package to obtain security vulnerability detection results, the source code in the workstation control exe software package is extracted to obtain the workstation control software source code, and the source code processing and analysis technology based on deep learning is used to perform sentence processing, semantic coding and optimization of the workstation control software source code, so as to automatically obtain the security vulnerability detection results according to the global correlation information of the semantic features of each optimized source code. In this way, potential vulnerabilities in the source code can be more accurately identified, including those problems that will only appear under specific runtime conditions. At the same time, complex business logic can be understood and analyzed, so as to discover logical errors hidden in deep conditional statements and program flows, thereby improving the intelligence of software vulnerability detection.

[0036] Figure 2The present invention is a flowchart of performing vulnerability detection on the workstation control exe software package to obtain security vulnerability detection results in the method for intelligently screening software information of motor vehicle inspection and testing equipment according to an embodiment of the present application. Figure 3 The following is a data flow diagram of performing vulnerability detection on the workstation control exe software package to obtain security vulnerability detection results in the intelligent screening method for software information of motor vehicle inspection and testing equipment according to an embodiment of the present application. Figure 2 and Figure 3 As shown, the vulnerability detection is performed on the workstation control exe software package to obtain a security vulnerability detection result, including: S141, extracting the source code in the workstation control exe software package to obtain the workstation control software source code; S142, performing sentence processing on the workstation control software source code to obtain a sequence of workstation control software source code description sentences; S143, semantically encoding each workstation control software source code description sentence in the sequence of workstation control software source code description sentences to obtain a sequence of workstation control software source code sentence granularity semantic encoding feature vectors; S144. 44. Perform feature association optimization processing on the sequence of sentence-granularity semantic coding feature vectors of the workstation control software source code based on the quantization of the sequence's intrinsic correlation to obtain a sequence of sentence-granularity semantic optimization coding feature vectors of the workstation control software source code; S145. Input the sequence of sentence-granularity semantic optimization coding feature vectors of the workstation control software source code into a code sentence granularity global semantic association encoder to obtain a global semantic coding feature vector of the workstation control software source code; S146. Based on the global semantic coding feature vector of the workstation control software source code, obtain the security vulnerability detection result.

[0037] In step S141, the source code in the station machine control exe software package is extracted to obtain the station machine control software source code. It should be understood that the source code is the cornerstone of the software, and it contains all the logic and instructions of the software. By extracting the source code, the operating mechanism and business process of the software can be deeply understood, which is the premise for carrying out security vulnerability detection. And starting directly from the source code, those potential vulnerabilities that are difficult to detect by conventional means after compilation can be captured. These potential vulnerabilities may be hidden in complex conditional statements, loop structures or specific function calls, which will pose a threat to the security of the software. In general, the station machine control software source code extracted is the direct data of security vulnerability detection. By analyzing it, security vulnerabilities can be effectively identified, which helps technical maintenance personnel to find and solve security problems in time, thereby improving the security and reliability of the software.

[0038] In step S142, the source code of the workstation control software is processed by sentence segmentation to obtain a sequence of description sentences of the source code of the workstation control software. Accordingly, considering that the source code of the workstation control software contains specific patterns, structures or function call information in different codes, these may be signs of security vulnerabilities. Based on this, in order to understand and analyze the logical structure of each source code description sentence more carefully, in the technical solution of the present application, the source code of the workstation control software is processed by sentence segmentation to obtain a sequence of description sentences of the source code of the workstation control software. That is, the sentence segmentation process decomposes the source code of the workstation control software into smaller, manageable units (description sentences), which helps to analyze the structure and logic of the code more accurately.

[0039] In step S143, semantic encoding is performed on each workstation control software source code description sentence in the sequence of the workstation control software source code description sentences to obtain a sequence of semantic encoding feature vectors of the workstation control software source code sentence granularity. Accordingly, considering that each of the workstation control software source code description sentences contains patterns and structures about specific codes, this is very important for discovering potential security vulnerabilities and code quality issues. Based on this, in the technical solution of the present application, semantic encoding is performed on each workstation control software source code description sentence in the sequence of the workstation control software source code description sentences to convert the code description sentences into information that can represent their semantics, and obtain a sequence of semantic encoding feature vectors of the workstation control software source code sentence granularity.

[0040] In step S144, the sequence of the granularity semantic coding feature vectors of the source code sentence of the workstation control software is subjected to a feature association optimization process based on the quantification of the sequence's intrinsic correlation to obtain a sequence of the granularity semantic optimization coding feature vectors of the source code sentence of the workstation control software. Specifically, in an embodiment of the present application, the sequence of the granularity semantic coding feature vectors of the source code sentence of the workstation control software is subjected to a feature association optimization process based on the quantification of the sequence's intrinsic correlation to obtain a sequence of the granularity semantic optimization coding feature vectors of the source code sentence of the workstation control software, including: inputting the sequence of the granularity semantic coding feature vectors of the source code sentence of the workstation control software into a feature context semantic association optimization module based on the quantification of the sequence's intrinsic correlation to obtain a sequence of the granularity semantic optimization coding feature vectors of the source code sentence of the workstation control software.

[0041] It should be understood that there are mutual relationships and contextual semantic information between the sequences of the granular semantic encoding feature vectors of the workstation control software source code, and each feature has different contributions and importance within the entire code semantic range. In order to enhance the expressive power of each feature vector so that each feature can more effectively convey its semantic information, in the technical solution of the present application, the sequence of the granular semantic encoding feature vectors of the workstation control software source code is input into a feature context semantic association optimization module based on sequence endogenous correlation quantification to obtain a sequence of granular semantic optimized encoding feature vectors of the workstation control software source code. In particular, the feature context semantic association optimization module based on sequence endogenous correlation quantification improves the accuracy and effectiveness of feature expression by quantifying and enhancing the semantic connection between feature vectors. Specifically, first, a set of granular semantic association score vectors of the workstation control software source code is obtained by calculating the semantic association relationship between any two feature vectors in the sequence of granular semantic encoding feature vectors of the workstation control software source code. It should be understood that the score vector helps to identify which parts of the code are important for interaction, so as to quantify the semantic connection between different code features and provide a basis for subsequent sequence semantic aggregation. Next, the set of obtained score vectors is averaged to capture the average semantic relevance of the entire code sequence to determine the global representation vector of the endogenous correlation of the sentence granularity sequence of the workstation control software source code. Then, based on the global representation vector of the endogenous correlation of the sentence granularity sequence of the workstation control software source code, the associated optimization factor of each workstation control software source code sentence granularity semantic encoding feature vector is calculated to enhance the discrimination between different feature vectors, so that each feature vector can more accurately represent its unique semantic information, thereby obtaining a sequence of semantic association optimization factors of the sentence granularity of the workstation control software source code. Furthermore, in order to enhance the discrimination between different optimization factors, especially when some factors are more important than other factors, the sequence of associated optimization factors is input into the Softmax activation function to convert them into weight factors for normalization processing, and a sequence of semantic association optimization weight factors of the sentence granularity of the workstation control software source code is obtained. Finally, based on each associated optimization weight factor, the sequence of the granularity semantic encoding feature vectors of the workstation control software source code sentence is weighted to enhance those features that are more important to the analysis objectives (such as vulnerability detection and code quality assessment), ensuring that the model pays more attention to these key information, thereby generating a sequence of granularity semantic optimization encoding feature vectors of the workstation control software source code sentence.

[0042] Specifically, Figure 4In the method for intelligently screening software information of motor vehicle inspection and testing equipment according to the embodiment of the present application, the sequence of the granular semantic encoding feature vectors of the source code sentence of the station control software is input into the feature context semantic association optimization module based on the quantification of the intrinsic correlation of the sequence to obtain the sequence of the granular semantic optimization encoding feature vectors of the source code sentence of the station control software. Figure 4 As shown, the sequence of the workstation control software source code sentence granularity semantic coding feature vectors is input into a feature context semantic association optimization module based on sequence intrinsic correlation quantification to obtain the sequence of the workstation control software source code sentence granularity semantic optimization coding feature vectors, including: S1441, calculating the semantic association score vector between any two workstation control software source code sentence granularity semantic coding feature vectors in the sequence of the workstation control software source code sentence granularity semantic coding feature vectors to obtain a set of workstation control software source code sentence granularity semantic association score vectors; S1442, based on the set of the workstation control software source code sentence granularity semantic association score vectors, determining the workstation control software source code sentence granularity sequence intrinsic correlation global representation vector; S1443, based on the workstation control software source code sentence granularity sequence intrinsic correlation global representation vector, calculating the workstation control software The associated optimization factors of each workstation control software source code sentence granularity semantic coding feature vector in the sequence of source code sentence granularity semantic coding feature vectors are used to obtain a sequence of workstation control software source code sentence granularity semantic association optimization factors; S1444, the sequence of workstation control software source code sentence granularity semantic association optimization factors is activated to obtain a sequence of workstation control software source code sentence granularity semantic association optimization weight factors; S1445, each workstation control software source code sentence granularity semantic association optimization weight factor in the sequence of workstation control software source code sentence granularity semantic association optimization weight factors is used as a weight, and each workstation control software source code sentence granularity semantic coding feature vector in the sequence of workstation control software source code sentence granularity semantic coding feature vectors is weighted respectively to obtain a sequence of workstation control software source code sentence granularity semantic optimization coding feature vectors.

[0043] More specifically, in an embodiment of the present application, the semantic association score vector between any two workstation control software source code sentence granularity semantic encoding feature vectors in the sequence of the workstation control software source code sentence granularity semantic encoding feature vectors is calculated to obtain a set of workstation control software source code sentence granularity semantic association score vectors, including: cascading any two workstation control software source code sentence granularity semantic encoding feature vectors in the sequence of the workstation control software source code sentence granularity semantic encoding feature vectors to obtain a set of workstation control software source code sentence granularity semantic cascade feature vectors; multiplying each workstation control software source code sentence granularity semantic cascade feature vector in the set of the workstation control software source code sentence granularity semantic cascade feature vectors with the workstation control software source code sentence granularity semantic weight matrix, and then adding the resultant vectors with the workstation control software source code sentence granularity semantic bias vector by position to obtain the set of the workstation control software source code sentence granularity semantic association score vectors.

[0044] More specifically, in an embodiment of the present application, based on a set of granularity semantic association score vectors of the workstation control software source code sentence, a global representation vector of the intrinsic correlation of the workstation control software source code sentence granularity sequence is determined, including: calculating the mean vector of the set of granularity semantic association score vectors of the workstation control software source code sentence as the global representation vector of the intrinsic correlation of the workstation control software source code sentence granularity sequence.

[0045] More specifically, in an embodiment of the present application, based on the global representation vector of the endogenous correlation of the workstation control software source code sentence granularity sequence, the associated optimization factor of each workstation control software source code sentence granularity semantic coding feature vector in the sequence of the workstation control software source code sentence granularity semantic coding feature vector is calculated to obtain a sequence of workstation control software source code sentence granularity semantic association optimization factors, including: multiplying the global representation vector of the endogenous correlation of the workstation control software source code sentence granularity sequence with the global weight matrix of the workstation control software source code to obtain the global representation weight vector of the workstation control software source code; multiplying the global representation vector of the endogenous correlation of the workstation control software source code sentence granularity sequence with the global weight matrix of the workstation control software source code to obtain the global representation weight vector of the workstation control software source code; multiplying the global representation vector of the endogenous correlation of the workstation control software source code sentence granularity semantic coding feature vector of each workstation control software source code sentence granularity semantic coding feature vector of The sentence granularity semantic encoding feature vector of the control software source code is multiplied by the local weight matrix of the workstation control software source code to obtain a set of local representation weight vectors of the workstation control software source code; each local representation weight vector of the workstation control software source code in the set of local representation weight vectors of the workstation control software source code is added to the global representation weight vector of the workstation control software source code and the bias vector of the workstation control software source code by position to obtain a set of sentence granularity semantic association vectors of the workstation control software source code; the set of sentence granularity semantic association vectors of the workstation control software source code is input into the Sigmoid function to obtain a sequence of sentence granularity semantic association optimization factors of the workstation control software source code.

[0046] More specifically, in an embodiment of the present application, the sequence of granularity semantic association optimization factors of the workstation control software source code sentence is activated to obtain a sequence of granularity semantic association optimization weight factors of the workstation control software source code sentence, including: inputting the sequence of granularity semantic association optimization factors of the workstation control software source code sentence into a Softmax activation function to obtain a sequence of granularity semantic association optimization weight factors of the workstation control software source code sentence.

[0047] In the embodiment of the present application, specifically, the sequence of the semantic coding feature vectors of the sentence granularity of the workstation control software source code is subjected to feature association optimization processing based on the quantification of the sequence's intrinsic correlation to obtain a sequence of semantic optimization coding feature vectors of the sentence granularity of the workstation control software source code, including: inputting the sequence of semantic coding feature vectors of the sentence granularity of the workstation control software source code into a feature context semantic association optimization module based on the quantification of the sequence's intrinsic correlation, and processing with the following semantic association optimization formula to obtain a sequence of semantic optimization coding feature vectors of the sentence granularity of the workstation control software source code; wherein the semantic association optimization formula is:

[0048]

[0049] in, is a sequence of granular semantic encoding feature vectors of the source code sentence of the workstation control software, The number of feature vectors in the sequence representing the granular semantic encoding feature vectors of the source code sentence of the workstation control software, and are respectively the first and The granular semantic encoding feature vector of the source code sentence of the workstation control software, For cascade processing operations, is the granularity semantic weight matrix of the source code sentence of the workstation control software, is the semantic bias vector of the source code sentence granularity of the workstation control software, is the first in the set of granular semantic association score vectors of the source code sentence of the workstation control software The semantic association score vector of the source code sentence of the workstation control software is is the number of vectors in the set of granular semantic association score vectors of the source code sentence of the workstation control software, is the global representation vector of the endogenous correlation of the granularity sequence of the source code sentence of the workstation control software, and are the global weight matrix of the workstation control software source code and the local weight matrix of the workstation control software source code, respectively. is the source code offset vector of the station machine control software, yes function, yes function The first in the sequence of weight factors for optimizing the granular semantic association of the source code sentence of the workstation control software The weight factor of the semantic association optimization of the source code sentence granularity of the workstation control software is The first in the sequence of the granular semantic optimization encoding feature vector of the source code sentence of the workstation control software The feature vector of sentence-granular semantic optimization encoding of source code of individual workstation machine control software.

[0050] In step S145, the sequence of the workstation control software source code sentence granularity semantic optimization encoding feature vectors is input into the code sentence granularity global semantic association encoder to obtain the workstation control software source code global semantic encoding feature vectors. Specifically, in an embodiment of the present application, the sequence of the workstation control software source code sentence granularity semantic optimization encoding feature vectors is input into the code sentence granularity global semantic association encoder to obtain the workstation control software source code global semantic encoding feature vectors, including: inputting the sequence of the workstation control software source code sentence granularity semantic optimization encoding feature vectors into the code sentence granularity global semantic association encoder based on the converter structure to obtain the workstation control software source code global semantic encoding feature vectors. It should be understood that each vector in the sequence of the granular semantic optimization encoding feature vectors of the workstation control software source code sentence represents the structure and grammatical information of a code sentence, and in order to effectively capture the global dependencies between different parts of the source code to understand the overall logic and structure of the code, in the technical solution of the present application, the sequence of the granular semantic optimization encoding feature vectors of the workstation control software source code sentence is input into the code sentence granular global semantic association encoder based on the converter structure to obtain the global semantic encoding feature vector of the workstation control software source code. That is, through the converter structure, the model can process and integrate long-distance dependencies and context information in the code, thereby enhancing the overall semantic understanding of the code semantics.

[0051] In step S146, the sequence of the semantic optimization encoding feature vectors of the station control software source code at the sentence granularity is input into the code sentence granularity global semantic association encoder to obtain the global semantic encoding feature vector of the station control software source code. Specifically, in the embodiment of the present application, based on the global semantic encoding feature vector of the station control software source code, the security vulnerability detection result is obtained, including: inputting the global semantic encoding feature vector of the station control software source code into a vulnerability detection module based on a classifier to obtain the security vulnerability detection result. That is, the global semantic encoding feature vector of the station control software source code obtained by semantically associating the sequence of the semantic optimization encoding feature vectors of the station control software source code at the sentence granularity is used for classification processing, so as to automatically obtain the security vulnerability detection result. In particular, the security vulnerability detection result here is whether there is a known security vulnerability. In this way, potential vulnerabilities in the source code can be more accurately identified, including those problems that will only appear under specific runtime conditions. At the same time, it can understand and analyze complex business logic, so as to discover logical errors hidden in deep conditional statements and program flows, thereby improving the intelligence of software vulnerability detection.

[0052] In particular, considering that the sequence of the station machine control software source code sentence granularity semantic coding feature vectors represents the semantic embedding coding features of each station machine control software source code description sentence determined by the station machine control software source code through sentence segmentation processing. When the sequence of the station machine control software source code sentence granularity semantic coding feature vectors is input into the feature context semantic association optimization module based on sequence intrinsic correlation quantification, the feature context semantic association optimization module based on sequence intrinsic correlation quantification optimizes the context semantic information expression richness of each station machine control software source code sentence granularity semantic coding feature vector in the sequence of the station machine control software source code sentence granularity semantic coding feature vectors by calculating the semantic association between each station machine control software source code sentence granularity semantic coding feature vector in the sequence of the station machine control software source code sentence granularity semantic coding feature vectors. However, this will also cause a certain amount of information expression overflow in each station machine control software source code sentence granularity semantic optimization coding feature vector in the sequence of the station machine control software source code sentence granularity semantic optimization coding feature vectors. Therefore, when the sequence of the sentence-granularity semantic optimization encoding feature vectors of the workstation control software source code is input into the code sentence-granularity global semantic association encoder based on the converter structure, the code sentence-granularity global semantic association encoder based on the converter structure captures the contextual semantic association information of the sequence of the sentence-granularity semantic optimization encoding feature vectors of the workstation control software source code in the global range through a multi-head attention mechanism. However, the superposition of the dual attention mechanisms will make the global semantic encoding feature vector of the workstation control software source code have a diversified set expression distribution of text semantic context aggregation features. Therefore, it is expected to improve the balance between the regression mapping accuracy and completeness of the global semantic encoding feature vector of the workstation control software source code when performing class regression through the classifier-based vulnerability detection module, thereby improving the accuracy of the obtained security vulnerability detection results.

[0053] In a preferred example, the global semantic coding feature vector of the workstation control software source code is passed through the classifier-based vulnerability detection module to obtain a security vulnerability detection result, including: determining the maximum eigenvalue of the global semantic coding feature vector of the workstation control software source code, the minimum eigenvalue of the global semantic coding feature vector of the workstation control software source code, and calculating the mean value of the global semantic coding feature vector of the workstation control software source code and the standard deviation of the global semantic coding feature vector of the workstation control software source code; calculating the quotient of the mean value of the global semantic coding feature vector of the workstation control software source code and the standard deviation of the global semantic coding feature vector of the workstation control software source code to obtain a statistical normalization value of the global semantic coding feature vector of the workstation control software source code; calculating the inverse of each eigenvalue of the global semantic coding feature vector of the workstation control software source code, and comparing it with the global semantic coding feature vector of the workstation control software source code. After multiplying the difference between the maximum eigenvalue of the encoding and the minimum eigenvalue of the global semantic encoding of the workstation control software source code, the result is subtracted from the statistical normalized value of the global semantic encoding of the workstation control software source code to obtain the global semantic encoding distribution approximation vector of the workstation control software source code; an exponential function with a natural constant as the base and each eigenvalue of the global semantic encoding distribution approximation vector of the workstation control software source code is calculated, and the result is added to the statistical normalized value of the global semantic encoding of the workstation control software source code to obtain the global semantic encoding probability approximation vector of the workstation control software source code; the logarithm with base two of the absolute value of each eigenvalue of the global semantic encoding probability approximation vector of the workstation control software source code is calculated to obtain the optimized global semantic encoding feature vector of the workstation control software source code; the optimized global semantic encoding feature vector of the workstation control software source code is passed through the classifier-based vulnerability detection module to obtain the security vulnerability detection result.

[0054] The optimized representation of the global semantic coding feature vector of the workstation control software source code is:

[0055]

[0056] in, is the global semantic encoding feature vector of the workstation control software source code, , , and They are the feature sets of the global semantic encoding feature vector of the workstation control software source code The mean, standard deviation, minimum and maximum values ​​of Indicates adding by position point. represents the natural exponential function, It means point multiplication by position, Logarithm to base 2, To calculate the inverse of each eigenvalue of the global semantic encoding eigenvector of the workstation control software source code, It means to subtract by position point. A global semantic encoding feature vector for the optimized workstation control software source code.

[0057] Therefore, by performing random statistical normalization on the diversified feature set of the global semantic coding feature vector of the workstation control software source code, the confidence space constructed based on the overall feature value of the global semantic coding feature vector of the workstation control software source code is approximated by the standardized continuous probability density distribution relative to the response hypothesis test of each feature value of the global semantic coding feature vector of the workstation control software source code, thereby establishing the target reachability of the discretized feature distribution of the global semantic coding feature vector of the workstation control software source code to the unified regression target, so as to achieve the balanced executableness between mapping accuracy and mapping completeness in the class regression process based on the discretized feature distribution of the global semantic coding feature vector of the workstation control software source code, and improve the accuracy of the security vulnerability detection results obtained by the global semantic coding feature vector of the workstation control software source code through the classifier-based vulnerability detection module. In this way, potential vulnerabilities in the source code can be more accurately identified, including those problems that will only appear under specific runtime conditions. At the same time, it can understand and analyze complex business logic, thereby discovering logical errors hidden in deep conditional statements and program flows, thereby improving the intelligence of software vulnerability detection.

[0058] In summary, the method for intelligent screening of software information of motor vehicle inspection and testing equipment based on the embodiment of the present application is explained, which extracts the source code in the workstation control exe software package to obtain the workstation control software source code, and uses source code processing and analysis technology based on deep learning to perform sentence processing, semantic coding and optimization of the workstation control software source code, so as to automatically obtain security vulnerability detection results based on the global correlation information of the semantic features of each optimized source code. In this way, potential vulnerabilities in the source code can be more accurately identified, including those problems that will only appear under specific runtime conditions. At the same time, it is possible to understand and analyze complex business logic, thereby discovering logical errors hidden in deep conditional statements and program flows, thereby improving the intelligence level of software vulnerability detection.

[0059] The above is only a preferred embodiment of the present application and does not constitute any form of limitation to the present application. Although the present application has been disclosed as a preferred embodiment as above, it is not intended to limit the present application. Any technician familiar with the profession can make some changes or modifications to equivalent embodiments of the technical contents disclosed above without departing from the scope of the technical solution of the present application. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present application without departing from the content of the technical solution of the present application still falls within the scope of the technical solution of the present application.

Claims

1. A method for intelligently screening software information of motor vehicle inspection and testing equipment, characterized in that: include: Determine the directory location and file location of the motor vehicle inspection and testing equipment software; Based on the directory location and the file location, extract the workstation control exe software package; Extract key information from the workstation control exe software package, the key information including version number, compilation date, digital signature and configuration file; Perform vulnerability detection on the workstation control exe software package to obtain a security vulnerability detection result; Among them, performing vulnerability detection on the workstation control exe software package to obtain security vulnerability detection results includes: Extract the source code in the workstation control exe software package to obtain the workstation control software source code; Sentence processing is performed on the workstation control software source code to obtain a sequence of description sentences of the workstation control software source code; Semantically encoding each workstation control software source code description sentence in the sequence of workstation control software source code description sentences to obtain a sequence of granular semantic encoding feature vectors of the workstation control software source code sentences; Performing feature association optimization processing based on sequence endogenous correlation quantization on the sequence of granular semantic coding feature vectors of the workstation control software source code sentence to obtain a sequence of granular semantic optimized coding feature vectors of the workstation control software source code sentence; Inputting the sequence of the station machine control software source code sentence granularity semantic optimization coding feature vectors into the code sentence granularity global semantic association encoder to obtain the station machine control software source code global semantic coding feature vector; Based on the global semantic coding feature vector of the workstation control software source code, the security vulnerability detection result is obtained; The sequence of the workstation control software source code sentence granularity semantic coding feature vectors is subjected to feature association optimization processing based on sequence endogenous correlation quantification to obtain a sequence of the workstation control software source code sentence granularity semantic optimization coding feature vectors, including: inputting the sequence of the workstation control software source code sentence granularity semantic coding feature vectors into a feature context semantic association optimization module based on sequence endogenous correlation quantification to obtain a sequence of the workstation control software source code sentence granularity semantic optimization coding feature vectors; Inputting the sequence of the granular semantic coding feature vectors of the workstation control software source code sentence into a feature context semantic association optimization module based on sequence endogenous correlation quantization to obtain the sequence of the granular semantic optimization coding feature vectors of the workstation control software source code sentence, including: Calculating the semantic association score vector between any two workstation control software source code sentence granularity semantic encoding feature vectors in the sequence of the workstation control software source code sentence granularity semantic encoding feature vectors to obtain a set of workstation control software source code sentence granularity semantic association score vectors; Determine a global representation vector of endogenous correlation of the granularity sequence of the source code sentence of the workstation control software based on the set of granularity semantic association score vectors of the source code sentence of the workstation control software; Based on the global representation vector of the endogenous correlation of the workstation control software source code sentence granularity sequence, calculating the association optimization factor of each workstation control software source code sentence granularity semantic encoding feature vector in the sequence of the workstation control software source code sentence granularity semantic encoding feature vector to obtain a sequence of workstation control software source code sentence granularity semantic association optimization factors; Activating the sequence of the granularity semantic association optimization factors of the workstation control software source code sentences to obtain a sequence of granularity semantic association optimization weight factors of the workstation control software source code sentences; Taking each workstation control software source code sentence granularity semantic association optimization weight factor in the sequence of the workstation control software source code sentence granularity semantic association optimization weight factors as the weight, each workstation control software source code sentence granularity semantic coding feature vector in the sequence of the workstation control software source code sentence granularity semantic coding feature vector is weighted respectively to obtain the sequence of the workstation control software source code sentence granularity semantic optimization coding feature vectors.

2. The method for intelligently screening software information of motor vehicle inspection and testing equipment according to claim 1, characterized in that: Calculating the semantic association score vector between any two workstation control software source code sentence granularity semantic encoding feature vectors in the sequence of the workstation control software source code sentence granularity semantic encoding feature vectors to obtain a set of workstation control software source code sentence granularity semantic association score vectors, including: Cascading any two of the workstation control software source code sentence granularity semantic coding feature vectors in the sequence of the workstation control software source code sentence granularity semantic coding feature vectors to obtain a set of workstation control software source code sentence granularity semantic cascade feature vectors; Each workstation control software source code sentence granularity semantic cascade feature vector in the set of the workstation control software source code sentence granularity semantic cascade feature vectors is multiplied by the workstation control software source code sentence granularity semantic weight matrix, and then the resultant is added positionally to the workstation control software source code sentence granularity semantic bias vector to obtain a set of the workstation control software source code sentence granularity semantic association score vectors.

3. The method for intelligently screening software information of motor vehicle inspection and testing equipment according to claim 2, characterized in that: Based on the set of granularity semantic association score vectors of the workstation control software source code sentence, determining the global representation vector of the endogenous correlation of the workstation control software source code sentence granularity sequence, including: calculating the mean vector of the set of granularity semantic association score vectors of the workstation control software source code sentence as the global representation vector of the endogenous correlation of the workstation control software source code sentence granularity sequence.

4. The method for intelligently screening software information of motor vehicle inspection and testing equipment according to claim 3 is characterized in that: Based on the global representation vector of the endogenous correlation of the workstation control software source code sentence granularity sequence, the correlation optimization factor of each workstation control software source code sentence granularity semantic encoding feature vector in the sequence of the workstation control software source code sentence granularity semantic encoding feature vector is calculated to obtain a sequence of workstation control software source code sentence granularity semantic correlation optimization factors, including: Multiplying the endogenous correlation global representation vector of the workstation control software source code sentence granularity sequence with the workstation control software source code global weight matrix to obtain a global representation weight vector of the workstation control software source code; Multiplying each station machine control software source code sentence granularity semantic coding feature vector in the sequence of station machine control software source code sentence granularity semantic coding feature vectors with the station machine control software source code local weight matrix to obtain a set of station machine control software source code local representation weight vectors; Each workstation control software source code local representation weight vector in the set of the workstation control software source code local representation weight vectors is added to the workstation control software source code global representation weight vector and the workstation control software source code bias vector by position to obtain a set of workstation control software source code sentence granularity semantic association vectors; The set of the granularity semantic association vectors of the source code sentence of the workstation control software is input into a Sigmoid function to obtain a sequence of granularity semantic association optimization factors of the source code sentence of the workstation control software.

5. The method for intelligently screening software information of motor vehicle inspection and testing equipment according to claim 4, characterized in that: The sequence of the granularity semantic association optimization factors of the workstation control software source code sentence is activated to obtain a sequence of the granularity semantic association optimization weight factors of the workstation control software source code sentence, including: inputting the sequence of the granularity semantic association optimization factors of the workstation control software source code sentence into a Softmax activation function to obtain a sequence of the granularity semantic association optimization weight factors of the workstation control software source code sentence.

6. The method for intelligently screening software information of motor vehicle inspection and testing equipment according to claim 5, characterized in that: Inputting the sequence of the workstation control software source code sentence granularity semantic optimization coding feature vectors into a code sentence granularity global semantic association encoder to obtain the workstation control software source code global semantic coding feature vector, including: inputting the sequence of the workstation control software source code sentence granularity semantic optimization coding feature vectors into a code sentence granularity global semantic association encoder based on a converter structure to obtain the workstation control software source code global semantic coding feature vector.

7. The method for intelligently screening software information of motor vehicle inspection and testing equipment according to claim 6, characterized in that: Based on the global semantic coding feature vector of the workstation control software source code, the security vulnerability detection result is obtained, including: inputting the global semantic coding feature vector of the workstation control software source code into a classifier-based vulnerability detection module to obtain the security vulnerability detection result.

Images