A detection method for open source component libraries of binary programs
By adjusting the weight value of open source components and calculating similarity, the problem of low recognition accuracy in the binary open source component library version detection is solved, and more accurate version recognition and distinction is achieved, which improves detection efficiency.
Patent Information
- Application Number
- CN202410914337.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-09
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-07-09
AI Technical Summary
The existing technology lacks features suitable for version matching in binary open source component library version detection, resulting in low recognition accuracy and the inability to accurately distinguish library versions of the same category.
By obtaining open source components of different categories and versions, extracting fingerprint features, and adjusting weight values based on these features, calculating the similarity between components to be classified and components of each category to achieve more accurate version recognition.
It improves the accuracy and distinction of open source component version recognition, improves the operation efficiency of the algorithm, avoids fingerprint feature extraction of all components, and enhances the ability to distinguish between components of the same category.
Smart Images

Figure CN118503791B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of open source component identification, and in particular to a detection method for an open source component library of a binary program. Background Art
[0002] Currently, software development widely relies on open source component libraries provided by the open source community to reuse existing software development results and improve development efficiency. While improving development efficiency, open source component libraries also expand the attack surface of software, bringing a series of security risks, such as potential vulnerabilities.
[0003] The main challenge facing binary open source component library version detection is that there are no features suitable for version matching, and the accuracy of library version identification cannot be improved. Accurate version identification is very important for research on vulnerability location and version updates, and the accuracy of identification depends largely on the completeness of the feature database and the ability of the selected features to identify the differences between libraries. Using only string frequency as a basis can distinguish different categories of libraries, but it cannot accurately distinguish libraries of the same category, so new features need to be added to expand the differences between library versions. Taking Scantist SCA as an example, it can scan binary files and identify the open source component libraries used and the related versions of the open source component libraries, but the accuracy of open source component library version identification is not high at present.
[0004] Based on this, the present invention is proposed. Summary of the invention
[0005] In view of the above problems, the present invention proposes a detection method for an open source component library of a binary program, and the method specifically comprises the following steps:
[0006] S1. Obtain multiple open source components of different categories and versions;
[0007] The open source component is a binary program file;
[0008] Each category of open source component collection includes multiple open source components of different versions;
[0009] S2. extract fingerprint features of each open source component from the multiple open source components;
[0010] S3. According to the fingerprint feature, adjust the weight value of each category of open source component set and the weight value corresponding to each open source component;
[0011] S4. Obtain the binary program file of the open source component to be classified;
[0012] S5. Extracting fingerprint features of the open source components to be classified according to the binary program file;
[0013] S6. Calculate the similarity between the open source component to be classified and each open source component in each category of the open source component set according to the adjusted weight value in step S3;
[0014] S7. Perform version identification on the open source component to be classified according to the similarity.
[0015] The beneficial effects of the present invention are as follows:
[0016] 1. The present invention improves the accuracy of version recognition of the open source components to be classified by selecting open source components in each open source component set to calculate the fingerprint features of the open source components, and adjusting the derivation function and string weight value of each fingerprint feature;
[0017] 2. The present invention improves the accuracy of open source component identification by screening selected open source components in the open source component set, and at the same time avoids the extraction of fingerprint features of all open source components, thereby improving the overall operating efficiency of the algorithm;
[0018] 3. The present invention improves the differentiation of open source components of different categories and the differentiation between open source components of the same category by setting two layers of weight values, namely, a weight value for a set of open source components and a weight value for each open source component, and dynamically updates the above weight values.
[0019] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above description and other purposes, features and advantages of the present invention more obvious and easy to understand, the preferred embodiments are specifically cited and described in detail as follows. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:
[0021] Figure 1 The flowchart of a detection method for binary open source component libraries is shown in FIG. DETAILED DESCRIPTION
[0022] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.
[0023] In the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be connected, detachably connected, or integrated; it can be mechanically connected or electrically connected; it can be directly connected or indirectly connected through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.
[0024] The present invention proposes a detection method for a binary open source component library, which specifically includes the following steps:
[0025] S1. Obtain multiple open source components of different categories and versions;
[0026] The open source component is a binary program file;
[0027] Each category of open source component collection includes multiple open source components of different versions;
[0028] S2. extract fingerprint features of each open source component from the multiple open source components;
[0029] S3. According to the fingerprint feature, adjust the weight value of each category of open source component set and the weight value corresponding to each open source component;
[0030] S4. Obtain the binary program file of the open source component to be classified;
[0031] S5. Extracting fingerprint features of the open source components to be classified according to the binary program file;
[0032] S6. Calculate the similarity between the open source component to be classified and each open source component in each category of the open source component set according to the adjusted weight value in step S3;
[0033] S7. Perform version identification on the open source component to be classified according to the similarity.
[0034] Furthermore, the step S1 specifically includes the following steps:
[0035] S11. Obtain multiple open source component collections of different categories from a preset open source component library { }, and obtain multiple different versions of open source components from each open source component collection; among them, is the open source component set of the ith category; n is the total number of categories of the open source component set in the open source component library;
[0036] Among them, the set of open source components of the same category includes multiple different versions of open source components;
[0037] The preset open source component library stores a plurality of different categories of open source component collections.
[0038] S12. For a collection of open source components Multiple different versions of open source components in { }, get the open source components in the middle ;
[0039] in, A collection of open source components The jth version of the open source component in m is the set of open source components. The number of versions in ; ;
[0040] Among them, open source components { }Sort by version update time from earliest to latest;
[0041] Among them, open source components { }The versions of various open source components are different; open source components Updated earlier than open source components The update time.
[0042] S13. Open source components The binary code in the binary program file is evenly divided into X parts to form X binary code intervals;
[0043] S14. Randomly obtain a binary code segment of a preset length L in each of the binary code intervals to form a binary code segment set { };in, is a binary code segment of a preset length L randomly obtained within the vth binary code interval;
[0044] S15. For open source components An open source component in , the open source components The binary code in the binary program file is evenly divided into X parts, forming X binary code intervals { };in and ; is the vth binary code interval; ;
[0045] S16. Traverse the binary code range , when the There is a binary code fragment in When the repetition value C is increased by 1; when the There is no binary code snippet in When , the repetition value C is not adjusted;
[0046] S17. For the binary code interval { }, execute step S16 for each binary code interval, and obtain a repetition value C after the execution is completed;
[0047] Computing open source components With open source components The binary similarity value of ;
[0048] S18. For open source components }, repeat steps S15 to S17 to obtain { Each open source component in} and the open source component The binary similarity value of { , };
[0049] Will{ , } Sort by value from small to large, select the first Y binary similarity values with the smallest value, and obtain multiple open source components corresponding to the first Y binary similarity values, and form the multiple open source components into this type of open source component set A collection of selected open source components ={ };
[0050] in, is the Yth selected open source component;
[0051] S19. Collection of open source components for multiple different categories }, execute steps S12 to S18 to obtain the selected open source component set corresponding to each type of open source component set { }.
[0052] Furthermore, the step S2 specifically includes the following steps:
[0053] S21. For each selected open source component corresponding to each type of open source component set, disassemble the open source component to obtain a disassembled result code;
[0054] Among them, tools for disassembling open source components include IDA tools, etc.
[0055] S22. Count the names and the number of times each derived function appears, and the names and the number of times each string appears in the disassembled result code as the fingerprint feature Z={(D,U),(F,W)} of the open source component;
[0056] Where, (D, U) = { };
[0057] (F, W) = { };
[0058] in, is the name of the pth exported function, is the number of times the pth export function appears; p is the total number of export functions counted after the open source component is disassembled;
[0059] is the name of the qth string, is the number of times the qth string appears; q is the total number of strings counted after the open source component is disassembled.
[0060] Furthermore, the step S3 specifically includes the following steps:
[0061] S301. Collection of open source components for different categories { A collection of open source components in , initialize the open source component collection The weight set of
[0062] Initialize the open source component collection The weight set specifically includes:
[0063] S3011. Get a collection of open source components The corresponding set of selected open source components ={ Fingerprint features of each open source component in};
[0064] S3012. For all the derived functions that appear in the fingerprint features of each open source component in step S3011, when one derived function exists in the fingerprint features of each open source component, the initial weight corresponding to the derived function is set to the first weight value; otherwise, the initial weight corresponding to the derived function is set to the second weight value;
[0065] Wherein, the first weight value is greater than the second weight value;
[0066] When a string exists in the fingerprint features of all open source components, the initial weight corresponding to the string is set to the third weight value; otherwise, the initial weight corresponding to the string is set to the fourth weight value;
[0067] Among them, the first weight value>the second weight value>the third weight value>the fourth weight value;
[0068] S3013. Use the initial weight values corresponding to each export function as an open source component set The corresponding initial weight set; the initial weight values corresponding to each export function and the initial weight values corresponding to each string in the open source component are used as the initial weight set of the open source component;
[0069] S302. Collection of open source components for different categories { }, execute step S301 to obtain the initial weight set corresponding to each open source component set as the latest weight set corresponding to each open source component set; use the initial weight set corresponding to each open source component in the selected open source component set corresponding to each open source component set as the latest weight set corresponding to each open source component; execute step S303;
[0070] S303. Collection of open source components for different categories A collection of open source components in , calculate the open source component set The corresponding set of selected open source components ={ Each open source component in} is relative to the open source component collection The associated value of
[0071] The associated value = ;
[0072] Where H is the number of exported functions in the open source component; is the latest weight value corresponding to the t-th export function in the open source component; is the first coefficient, when the open source component set When the derived function exists in the corresponding latest weight set and the weight value of the derived function exists, Set to 1, otherwise Set to 0;
[0073] Take each open source component relative to the open source component set The minimum value of the associated value;
[0074] S304. Computing open source component collection ={ }, each open source component relative to each open source component set { }, and take the maximum value of all the correlation values calculated in step S304;
[0075] S305. When the minimum value in step S303 is greater than or equal to the maximum value in step S304, executing step S306;
[0076] When the minimum value in step S303 is less than the maximum value in step S304, the open source component set is adjusted The weight values corresponding to each derived function in the corresponding latest weight set, and repeatedly executing step S303 to step S304 until the minimum value is greater than or equal to the maximum value, and then executing step S306;
[0077] S306. Computing open source component collection ={ }The score value of each open source component in
[0078] The rating value = ;
[0079] in, is the number of strings in the open source component; For open source components The latest weight value corresponding to the string;
[0080] S307. When open source components are collected ={ }When the difference between the score values of any two open source components in the system is less than the preset threshold value G, adjust the latest weight values of the strings in the two open source components until the difference between the score values of any two open source components is greater than or equal to the preset threshold value G;
[0081] Among them, the preset threshold G can be flexibly set according to actual experience.
[0082] The weight values corresponding to the export functions adjusted in step S305 are used as the open source component set. The corresponding latest weight value; the weight value corresponding to each derived function after adjustment in step S305 and the weight value corresponding to each string after adjustment in step S307 are used as the latest weight value corresponding to each open source component;
[0083] S308. Collection of open source components for different categories }, repeat steps S303 to S307 to obtain the latest weight value corresponding to each open source component set and the latest weight value corresponding to each selected open source component in each open source component set.
[0084] Furthermore, the step S5 specifically includes:
[0085] Extracting fingerprint features of the open source component to be classified according to the binary program file;
[0086] The fingerprint features include the names and occurrence times of each export function corresponding to the open source component to be classified, and the names and occurrence times of each string.
[0087] Furthermore, the step S6 specifically includes the following steps:
[0088] S61. Calculate the open source components to be classified and the set of each open source component { }, and the open source component set corresponding to the largest correlation value As the category of the open source component to be classified;
[0089] in, Indicates the open source component set corresponding to the maximum correlation value;
[0090] S62. Get a collection of open source components The corresponding set of selected open source components , and obtain a collection of open source components The latest weight value corresponding to each open source component in;
[0091] S63. Calculate the similarity between the open source component to be classified and each open source component according to the latest weight value corresponding to each open source component obtained in step S62;
[0092] Among them, similarity = ;
[0093] in, is the number of exported functions in the open source components to be classified; is the number of strings in the open source component to be classified; the score value is the score value of the open source component whose similarity is calculated with the open source component to be classified;
[0094] It is the first open source component to be classified The latest weight value of the derived function is: when the derived function exists in the fingerprint feature of the open source component whose similarity is calculated with the open source component to be classified, is equal to the latest weight value corresponding to the derived function in the open source component whose similarity is calculated with the open source component to be classified; when the derived function does not exist in the fingerprint feature of the open source component whose similarity is calculated with the open source component to be classified, is equal to 0;
[0095] It is the first open source component to be classified The latest weight value of a string is: when the string exists in the fingerprint feature of the open source component whose similarity is calculated with the open source component to be classified, is equal to the latest weight value corresponding to the string in the open source component whose similarity is calculated with the open source component to be classified; when the string does not exist in the fingerprint feature of the open source component whose similarity is calculated with the open source component to be classified, is equal to 0.
[0096] Furthermore, the step S7 specifically includes the following steps:
[0097] The minimum value of the multiple similarities obtained in step S63 is obtained, and the version of the open source component corresponding to the minimum value is used as the version of the open source component to be classified.
[0098] Furthermore, the step S7 further includes:
[0099] Obtain the minimum value of the multiple similarities obtained in step S63, and when the minimum value is greater than or equal to a preset threshold Feedback to the user the fingerprint features of the open source components to be classified, the binary files of each open source component corresponding to the open source component set of the category to which the open source components to be classified belong, and the fingerprint features of each open source component in the selected open source component set corresponding to the open source component set of the category to which the open source components to be classified belong;
[0100] When the minimum value is less than the preset threshold When the version of the open source component corresponding to the minimum value is used as the version of the open source component to be classified.
[0101] The preset threshold It can be set flexibly according to actual experience.
[0102] The beneficial effects of the present invention are as follows:
[0103] 1. The present invention improves the accuracy of version recognition of the open source components to be classified by selecting open source components in each open source component set to calculate the fingerprint features of the open source components, and adjusting the derivation function and string weight value of each fingerprint feature;
[0104] 2. The present invention improves the accuracy of open source component identification by screening selected open source components in the open source component set, and at the same time avoids the extraction of fingerprint features of all open source components, thereby improving the overall operating efficiency of the algorithm;
[0105] 3. The present invention improves the differentiation of open source components of different categories and the differentiation between open source components of the same category by setting two layers of weight values, namely, a weight value for a set of open source components and a weight value for each open source component, and dynamically updates the above weight values.
[0106] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.
Claims
1. A detection method for an open source component library of a binary program, characterized in that: The method specifically comprises the following steps: S1. Obtain multiple open source components of different categories and versions; The open source component is a binary program file; Each category of open source component collection contains multiple open source components of different versions; The step S1 specifically includes the following steps: S11. Obtain multiple open source component collections of different categories from a preset open source component library { }, and obtain multiple different versions of open source components from each open source component collection; among them, is the open source component set of the ith category; n is the total number of categories of the open source component set in the open source component library; Among them, the set of open source components of the same category includes multiple different versions of open source components; S12. For a collection of open source components Multiple different versions of open source components in { }, get the open source components in the middle ; in, A collection of open source components The jth version of the open source component in; m is the set of open source components The number of versions in ; where ; Among them, open source components { }Sort by version update time from earliest to latest; S13. Open source components The binary code in the binary program file is evenly divided into X parts to form X binary code intervals; S14. Randomly obtain a binary code segment of a preset length L in each of the binary code intervals to form a binary code segment set { };in, is a binary code segment of a preset length L randomly obtained within the vth binary code interval; S15. For open source components An open source component in , the open source components The binary code in the binary program file is evenly divided into X parts, forming X binary code intervals { };in and ; is the vth binary code interval; ; S16. Traverse the binary code range , when the There is a binary code fragment in When the repetition value C is increased by 1; when the There is no binary code snippet in When , the repetition value C is not adjusted; S17. For the binary code interval { }, execute step S16 for each binary code interval, and obtain a repetition value C after the execution is completed; Computing open source components With open source components The binary similarity value of ; S18. For open source components }, repeat steps S15 to S17 to obtain { Each open source component in} and the open source component The binary similarity value of { , }; Will{ , } Sort by value from small to large, select the first Y binary similarity values with the smallest value, and obtain multiple open source components corresponding to the first Y binary similarity values, and form the multiple open source components into an open source component set of this category A collection of selected open source components ={ }; in, is the Yth selected open source component; S19. Collection of open source components for multiple different categories }, execute steps S12 to S18 to obtain the selected open source component set corresponding to each type of open source component set { }; S2. extract fingerprint features of each open source component from the multiple open source components; S3. According to the fingerprint feature, adjust the weight value of each category of open source component set and the weight value corresponding to each open source component; S4. Obtain the binary program file of the open source component to be classified; S5. Extracting fingerprint features of the open source components to be classified according to the binary program file; S6. Calculate the similarity between the open source component to be classified and each open source component in each category of the open source component set according to the adjusted weight value in step S3; S7. Perform version identification on the open source component to be classified according to the similarity.
2. A detection method for an open source component library for binary programs according to claim 1, characterized in that: The step S2 specifically includes the following steps: S21. For each selected open source component corresponding to each type of open source component set, disassemble the open source component to obtain a disassembled result code; S22. Count the names and the number of times each derived function appears, and the names and the number of times each string appears in the disassembled result code as the fingerprint feature Z={(D,U),(F,W)} of the open source component; Where, (D, U) = { }; (F,W)={ }; in, is the name of the pth exported function, is the number of times the pth export function appears; p is the total number of export functions counted after the open source component is disassembled; is the name of the qth string, is the number of times the qth string appears; q is the total number of strings counted after the open source component is disassembled.
3. A detection method for an open source component library for binary programs according to claim 2, characterized in that: The step S3 specifically includes the following steps: S301. Collection of open source components for different categories { A collection of open source components in , initialize the open source component collection The weight set of Initialize the open source component collection The weight set specifically includes: S3011. Get a collection of open source components The corresponding set of selected open source components ={ Fingerprint features of each open source component in}; S3012. For all the derived functions that appear in the fingerprint features of each open source component in step S3011, when one derived function exists in the fingerprint features of each open source component, the initial weight corresponding to the derived function is set to the first weight value; otherwise, the initial weight corresponding to the derived function is set to the second weight value; Wherein, the first weight value is greater than the second weight value; When a string exists in the fingerprint features of all open source components, the initial weight corresponding to the string is set to the third weight value; otherwise, the initial weight corresponding to the string is set to the fourth weight value; Among them, the first weight value>the second weight value>the third weight value>the fourth weight value; S3013. Use the initial weight values corresponding to each export function as an open source component set The corresponding initial weight set; the initial weight values corresponding to each export function and the initial weight values corresponding to each string in the open source component are used as the initial weight set of the open source component; S302. Collection of open source components for different categories { }, execute step S301 to obtain the initial weight set corresponding to each open source component set as the latest weight set corresponding to each open source component set; use the initial weight set corresponding to each open source component in the selected open source component set corresponding to each open source component set as the latest weight set corresponding to each open source component; execute step S303; S303. Collection of open source components for different categories { A collection of open source components in , calculate the open source component set The corresponding set of selected open source components ={ Each open source component in} is relative to the open source component collection The associated value of The associated value = ; Where H is the number of exported functions in the open source component; is the latest weight value corresponding to the t-th export function in the open source component; is the first coefficient, when the open source component set When the derived function exists in the corresponding latest weight set and the weight value of the derived function exists, Set to 1, otherwise Set to 0; Take each open source component relative to the open source component set The minimum value of the associated value; S304. Computing open source component collection ={ }, each open source component relative to each open source component set { }, and take the maximum value of all the correlation values calculated in step S304; S305. When the minimum value in step S303 is greater than or equal to the maximum value in step S304, executing step S306; When the minimum value in step S303 is less than the maximum value in step S304, the open source component set is adjusted The weight values corresponding to each derived function in the corresponding latest weight set, and repeatedly executing step S303 to step S304 until the minimum value is greater than or equal to the maximum value, and then executing step S306; S306. Computing open source component collection ={ }The score value of each open source component in The rating value = ; in, is the number of strings in the open source component; For open source components The latest weight value corresponding to the string; S307. When open source components are collected ={ }When the difference between the score values of any two open source components in the system is less than the preset threshold value G, adjust the latest weight values of the strings in the two open source components until the difference between the score values of any two open source components is greater than or equal to the preset threshold value G; The weight values corresponding to the export functions adjusted in step S305 are used as the open source component set. The corresponding latest weight value; the weight value corresponding to each derived function after adjustment in step S305 and the weight value corresponding to each string after adjustment in step S307 are used as the latest weight value corresponding to each open source component; S308. Collection of open source components for different categories }, repeat steps S303 to S307 to obtain the latest weight value corresponding to each open source component set and the latest weight value corresponding to each selected open source component in each open source component set.
4. A detection method for an open source component library for binary programs according to claim 3, characterized in that: The step S5 specifically includes: Extracting fingerprint features of the open source component to be classified according to the binary program file; The fingerprint features include the names and occurrence times of each export function corresponding to the open source component to be classified, and the names and occurrence times of each string.
5. A detection method for an open source component library for binary programs according to claim 4, characterized in that: The step S6 specifically includes the following steps: S61. Calculate the open source components to be classified and the set of each open source component { }, and the open source component set corresponding to the largest correlation value As the category of the open source component to be classified; in, Indicates the open source component set corresponding to the maximum correlation value; S62. Get a collection of open source components The corresponding set of selected open source components , and obtain a collection of open source components The latest weight value corresponding to each open source component in; S63. Calculate the similarity between the open source component to be classified and each open source component according to the latest weight value corresponding to each open source component obtained in step S62; Among them, similarity = ; in, is the number of exported functions in the open source components to be classified; is the number of strings in the open source component to be classified; the score value is the score value of the open source component whose similarity is calculated with the open source component to be classified; It is the first open source component to be classified The latest weight value of the derived function is: when the derived function exists in the fingerprint feature of the open source component whose similarity is calculated with the open source component to be classified, is equal to the latest weight value corresponding to the derived function in the open source component whose similarity is calculated with the open source component to be classified; when the derived function does not exist in the fingerprint feature of the open source component whose similarity is calculated with the open source component to be classified, is equal to 0; It is the first open source component to be classified The latest weight value of a string is: when the string exists in the fingerprint feature of the open source component whose similarity is calculated with the open source component to be classified, is equal to the latest weight value corresponding to the string in the open source component whose similarity is calculated with the open source component to be classified; when the string does not exist in the fingerprint feature of the open source component whose similarity is calculated with the open source component to be classified, is equal to 0.
6. A detection method for an open source component library for binary programs according to claim 5, characterized in that: The step S7 specifically includes the following steps: The minimum value of the multiple similarities obtained in step S63 is obtained, and the version of the open source component corresponding to the minimum value is used as the version of the open source component to be classified.
7. A detection method for an open source component library for binary programs according to claim 6, characterized in that: The step S7 further comprises: Obtain the minimum value of the multiple similarities obtained in step S63, and when the minimum value is greater than or equal to a preset threshold Feedback to the user the fingerprint features of the open source components to be classified, the binary files of each open source component corresponding to the open source component set of the category to which the open source components to be classified belong, and the fingerprint features of each open source component in the selected open source component set corresponding to the open source component set of the category to which the open source components to be classified belong; When the minimum value is less than the preset threshold When the version of the open source component corresponding to the minimum value is used as the version of the open source component to be classified.
