Computer remote login recognition system and method based on artificial intelligence

Abstract

The present invention relates to the technical field of computer remote login identification, and in particular to an artificial intelligence-based computer remote login identification system and method, comprising a login platform, wherein the login platform is internally provided with a server, a self-checking and verification unit, an operation evaluation unit, an encryption unit, a security analysis unit, a database, and a display unit; the present invention analyzes verification data in an in-depth and feedback manner to improve the security of computational login identification and reduce the risk rate of information loss; at the same time, through data analysis and feedback, it is determined whether the computer has been misused to improve the rigor of identification verification and ensure the security of the computer's internal data; and when analyzing real-time user facial feature images, hierarchical analysis and comparison are performed to ensure the integrity and effectiveness of data collection of identification cameras, which helps to improve login identification efficiency and improve the supervision and early warning performance of identification cameras.

Description

Technical Field

[0001] The present invention relates to the technical field of computer remote login recognition, and in particular to a computer remote login recognition system and method based on artificial intelligence. Background Art

[0002] A modern electronic computer is a type of electronic computing machine used for high-speed calculations. It can perform numerical and logical calculations, has storage and memory functions, can run according to programs, and can automatically and quickly process large amounts of data. It consists of hardware and software systems. Computers are used in the control systems of various industrial and consumer devices, including simple special-purpose devices, industrial equipment, and general-purpose equipment.

[0003] With the continuous progress of society and the continuous development of various hardware technologies, biometric recognition technology has also been widely developed. Biometric recognition technologies include fingerprint recognition, iris recognition, etc., one of which is face recognition. The current computer remote login system also uses passwords. However, the security level is too low, there is a risk of computer login by impersonation, which in turn causes data loss, and it is impossible to provide timely warning processing for high password error rates. At the same time, for face recognition, it is impossible to supervise whether the recognition camera is operating normally, resulting in low recognition efficiency or even failure to recognize, reducing the efficiency of computer login recognition;

[0004] In view of the above technical defects, a solution is now proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide a computer remote login identification system and method based on artificial intelligence to solve the technical defects mentioned above. The present invention collects user verification data and analyzes the verification data in an in-depth and feedback manner to improve the security of computational login identification and reduce the risk of information loss. At the same time, through data analysis and feedback, it determines whether the computer has been stolen to improve the security supervision of the computer, and at the same time improves the rigor of identification verification to ensure the security of the computer's internal data. When analyzing the real-time user facial feature image, the operating data of the identification camera is collected and compared in a hierarchical manner to ensure the integrity and effectiveness of the data collection of the identification camera, which helps to improve the login identification efficiency and improve the supervision and early warning performance of the identification camera.

[0006] The object of the present invention can be achieved by the following technical solutions: a computer remote login recognition system based on artificial intelligence, comprising a login platform, wherein the login platform is internally provided with a server, a self-checking and verification unit, an operation evaluation unit, an encryption unit, a security analysis unit, a database, and a display unit;

[0007] When the server in the login platform generates a login instruction and sends the login instruction to the self-test verification unit, the self-test verification unit immediately collects the user's verification data after receiving the login instruction. The verification data includes a real-time user facial feature image, a real-time input password, and the user's ID number, and analyzes the verification data. The unauthorized signal and the security signal obtained are sent to the display unit, the feedback signal is sent to the operation evaluation unit, and the risk signal obtained is sent to the encryption unit and the security analysis unit;

[0008] After receiving the feedback signal, the operation evaluation unit immediately collects the operation data of the recognition camera, including the ambient humidity value of the recognition camera, the dust volume on the mirror surface, and the line operating current, analyzes the operation data, sends the obtained unqualified signal to the display unit, and sends the obtained management signal to the display unit and the self-test verification unit;

[0009] After receiving the risk signal, the security analysis unit immediately obtains the real-time password input by the registrant within the time threshold, performs a security risk assessment analysis on the real-time password input by the registrant, obtains a comparison instruction and input signal, and sends the obtained input signal to the self-checking and verification unit;

[0010] When the encryption unit receives a risk signal and a high-risk signal, it obtains a self-locking signal and sends the self-locking signal to the display unit through the self-test verification unit. After receiving the self-locking signal, the display unit immediately displays the text "Login identification abnormality" and stops password verification at the same time.

[0011] Preferably, the verification data analysis process of the self-check verification unit is as follows:

[0012] Step 1: The time interval after the server sends the login command is collected and marked as the time threshold. The user ID number and real-time user facial feature image of the login user within the time threshold are obtained. At the same time, the facial feature image corresponding to the login user's user ID number is retrieved from the database and marked as the verification facial feature image;

[0013] Step 2: Compare and analyze the user ID number and real-time user facial feature image with the internally stored preset user ID number list and verification facial feature image:

[0014] If the user ID number is not in the preset user ID number list and the real-time user facial feature image is not equal to the verification facial feature image, or if the user ID number is not in the preset user ID number list and the real-time user facial feature image is equal to the verification facial feature image, an unauthorized signal is generated;

[0015] If the user ID number is within the preset user ID number list and the real-time user facial feature image is not equal to the verification facial feature image, a feedback signal is generated;

[0016] If the user's ID number is within the preset user ID number list and the real-time user facial feature image is equal to the verification facial feature image, a secondary verification signal is generated.

[0017] Preferably, when the secondary verification signal is generated in the self-check verification unit, the real-time input password of the login user within the time threshold is immediately obtained, and the real-time input password is compared and analyzed with the preset password stored in the internal input:

[0018] If the real-time input password is equal to the preset password, a security signal is generated;

[0019] If the real-time password is not equal to the preset password, a risk signal is generated.

[0020] Preferably, the operation data analysis process of the operation evaluation unit is as follows:

[0021] SS1: Obtain the ambient humidity value HS for the camera within the time threshold, and the dust volume JF on the mirror of the camera within the time threshold, and calculate the interference recognition value GS according to the formula;

[0022] SS12: Divide the time threshold into i sub-time nodes, where i is a natural number greater than zero. Obtain the line operating current of the identification camera in each sub-time node, construct a set A of line operating currents, and obtain the discrete values ​​of set A. These values ​​are marked as discrete risk values. The maximum and minimum subsets of set A are also obtained, and the difference between the maximum and minimum subsets is marked as the maximum span value. The product of the discrete risk value and the maximum span value is marked as the fault risk value GF.

[0023] SS13: Obtain the operating status coefficient H according to the formula, and compare and analyze the operating status coefficient H with the preset operating status coefficient threshold stored internally:

[0024] If the operating state coefficient H is less than or equal to the preset operating state coefficient threshold, a failure signal is generated;

[0025] If the operating state coefficient H is greater than a preset operating state coefficient threshold, a management signal is generated.

[0026] Preferably, the security risk assessment analysis process of the security analysis unit is as follows:

[0027] The real-time password input by the user within the time threshold is obtained, and the digital, alphabetic, and symbol features of the real-time password are extracted. The string of extracted numbers, letters, and symbols is marked as the verification feature character. The total number of verification feature characters is obtained and marked as the verification number. The verification number is compared and analyzed with the preset verification number threshold:

[0028] If the verification number is not equal to the preset verification number threshold, an input signal is generated;

[0029] If the verification number is equal to the preset verification number threshold, a comparison instruction is generated.

[0030] Preferably, when the security analysis unit generates a comparison instruction, it compares the verification feature characters with the preset verification feature characters stored in the internal storage one by one:

[0031] The number of numbers, letters, and symbols that are identical to the preset verification feature characters is obtained and marked as the correct number. The number of numbers, letters, and symbols that are different from the preset verification feature characters is obtained and marked as the incorrect number. The ratio of the correct number to the incorrect number is obtained and marked as the safety factor ratio. The safety factor ratio is compared and analyzed with the preset safety factor ratio threshold value stored internally:

[0032] If the safety factor ratio is less than the preset safety factor ratio threshold, a high risk signal is generated;

[0033] If the safety factor ratio is greater than or equal to the preset safety factor ratio threshold, a pop-up signal is generated.

[0034] The beneficial effects of the present invention are as follows:

[0035] (1) The present invention collects user verification data and analyzes the verification data in an in-depth and feedback manner to improve the security of computer-based login identification and reduce the risk of information loss. At the same time, the analysis range is improved by maintaining the real-time user facial feature image, real-time input password and user ID number, which helps to improve the accuracy and comprehensiveness of the analysis results. At the same time, a security risk assessment analysis is performed on the real-time input password of the registrant through data analysis and feedback to determine whether the computer has been stolen, thereby improving the security supervision effect of the computer and improving the rigor of identification verification to ensure the security of the computer's internal data;

[0036] (2) When analyzing the real-time user facial feature image, the present invention collects the operating data of the recognition camera and performs hierarchical analysis and comparison to ensure the integrity and effectiveness of the data collection of the recognition camera, which helps to improve the login recognition efficiency and improve the supervision and early warning performance of the recognition camera. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The present invention will be further described below with reference to the accompanying drawings;

[0038] Figure 1 It is a flow chart of the system of the present invention;

[0039] Figure 2 It is a local analysis diagram of the present invention. DETAILED DESCRIPTION

[0040] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0041] Example 1:

[0042] See also Figure 1-2 As shown, the present invention is a computer remote login and identification system based on artificial intelligence, including a login platform, wherein a server, a self-test verification unit, an operation evaluation unit, an encryption unit, a security analysis unit, a database, and a display unit are provided inside the login platform, the server is connected to the self-test verification unit in a one-way communication connection, the database is connected to the self-test verification unit in a one-way communication connection, the self-test verification unit is connected to the operation evaluation unit in a two-way communication connection, the self-test verification unit and the operation evaluation unit are both connected to the display unit in a one-way communication connection, and the self-test verification unit is connected to the security analysis unit and the encryption unit in a two-way communication connection;

[0043] When the server in the login platform generates a login instruction and sends it to the self-checking and verification unit, the self-checking and verification unit immediately collects the user's verification data after receiving the login instruction. The verification data includes a real-time user facial feature image, a real-time input password, and the user's ID number. The unit analyzes the verification data to ensure the security of computer use and conducts supervision and early warning analysis on the verification data. The specific verification data supervision and early warning analysis process is as follows:

[0044] The time duration after the server sends the login instruction is collected and marked as the time threshold. The user ID number and real-time user facial feature image of the user logging in within the time threshold are obtained. At the same time, the facial feature image corresponding to the user ID number of the user logging in is retrieved from the database and marked as the verification facial feature image. The user ID number and real-time user facial feature image are compared and analyzed with the preset user ID number list and verification facial feature image stored in the database:

[0045] If the user ID number is not in the preset user ID number list, the real-time user facial feature image is not equal to the verification facial feature image, or if the user ID number is not in the preset user ID number list, the real-time user facial feature image is equal to the verification facial feature image, an unauthorized signal is generated and sent to the display unit. After receiving the unauthorized signal, the display unit immediately displays the text "unauthorized operation", thereby improving the security of computer-based login identification and reducing the risk of information loss;

[0046] If the user ID number is within the preset user ID number list and the real-time user facial feature image is not equal to the verification facial feature image, a feedback signal is generated and sent to the operation evaluation unit;

[0047] If the user's ID number is in the preset user ID number list, and the real-time user facial feature image is equal to the verification facial feature image, a secondary verification signal is generated. When the secondary verification signal is generated, the real-time password input by the user within the time threshold is immediately obtained, and the real-time password is compared with the preset password stored internally for analysis:

[0048] If the real-time input password is equal to the preset password, a security signal is generated and sent to the display unit. After receiving the security signal, the display unit immediately displays the text "authorized" to remind the login computer to operate and complete the remote login identification operation of the computer;

[0049] If the real-time input password is not equal to the preset password, a risk signal is generated and sent to the encryption unit and the security analysis unit.

[0050] Example 2:

[0051] After receiving the feedback signal, the operation evaluation unit immediately collects the operation data of the recognition camera. The operation data includes the ambient humidity value of the recognition camera, the dust volume on the mirror surface, and the line operating current. The operation data is analyzed to ensure the integrity and effectiveness of the data collection of the recognition camera, which helps to improve the login recognition efficiency. The specific operation data analysis process is as follows:

[0052] Get the ambient humidity value HS of the camera within the time threshold, and get the dust volume JF of the mirror of the camera within the time threshold, and then calculate the value through the formula Obtain the interference recognition value, where a1 and a2 are the preset proportional factor coefficients of the ambient humidity value and the mirror dust volume, respectively. The proportional factor coefficient is used to correct the deviation of various parameters in the formula calculation process, so as to make the calculation more accurate and the parameter data more accurate. a3 is the preset correction factor coefficient, which is 1.836. a1 and a2 are both positive numbers greater than zero. GS is the interference recognition value. It should be noted that the larger the value of the interference recognition value GS is, the greater the interference to the normal recognition efficiency of the recognition camera is, and the clarity of the facial feature image captured by the recognition camera is affected.

[0053] The time threshold is divided into i sub-time nodes, where i is a natural number greater than zero. The line operating current of the identification camera in each sub-time node is obtained, and a set A of line operating currents is constructed. The discrete value of set A is obtained and marked as a discrete risk value. It should be noted that the larger the value of the discrete risk value, the more unstable the line operating current is, and the greater the risk of abnormal operation of the identification camera. At the same time, the maximum subset and the minimum subset in set A are obtained, and the difference between the maximum subset and the minimum subset is marked as the maximum current span value. The product of the discrete risk value and the maximum current span value is marked as the fault risk value, labeled GF.

[0054] And through the formula The operating status coefficient is obtained, where b1 and b2 are the preset weight coefficients of the interference identification value and the fault risk value, respectively. Both b1 and b2 are positive numbers greater than zero. b3 is the preset correction coefficient, which is 1.932. H is the operating status coefficient. The operating status coefficient H is compared with the preset operating status coefficient threshold stored internally for analysis:

[0055] If the operating status coefficient H is less than or equal to the preset operating status coefficient threshold, a failure signal is generated and sent to the display unit. After receiving the failure signal, the display unit immediately displays the text "unauthorized operation", thereby improving the security of the calculation-based login recognition and improving the supervision and early warning performance of the recognition camera;

[0056] If the operating status coefficient H is greater than the preset operating status coefficient threshold, a management signal is generated and sent to the display unit and the self-test verification unit. After receiving the management signal, the display unit immediately displays the text "Identification Camera Management", which helps to improve the computer's login recognition rate and reduce the probability of failure risk of the recognition camera. In addition, after receiving the management signal, the self-test verification unit obtains the real-time user facial feature image after the management operation, and then performs a verification operation on the real-time user facial feature image after the management operation again to ensure the security and rigor of the login recognition, which helps to improve the accuracy of the analysis results.

[0057] Example 3:

[0058] After receiving the risk signal, the security analysis unit immediately obtains the real-time password input by the registrant within the time threshold, and performs a security risk assessment analysis on the real-time password input by the registrant to determine whether the error rate of the real-time password input by the registrant is too high, and then determines whether the computer has been stolen, so as to improve the security supervision effect of the computer. The specific security risk assessment and analysis process is as follows:

[0059] The real-time password input by the user within the time threshold is obtained, and the digital, alphabetic, and symbol features of the real-time password are extracted. The string of extracted numbers, letters, and symbols is marked as the verification feature character. The total number of verification feature characters is obtained and marked as the verification number. The verification number is compared and analyzed with the preset verification number threshold:

[0060] If the number of verifications is not equal to the preset verification number threshold, an input signal is generated and sent to the self-checking and verification unit. After receiving the input signal, the self-checking and verification unit obtains the second password input by the registrant and performs password verification again to improve the rigor of identification and verification;

[0061] If the verification number is equal to the preset verification number threshold, a comparison instruction is generated. When the comparison instruction is generated, the verification feature characters are compared one by one with the preset verification feature characters stored in the internal storage:

[0062] The number of numbers, letters, and symbols corresponding to the verification feature characters that are the same as the preset verification feature characters is obtained, and marked as the correct number; the number of numbers, letters, and symbols corresponding to the verification feature characters that are different from the preset verification feature characters is obtained, and marked as the wrong number; the ratio of the correct number to the wrong number is obtained, and marked as the security factor ratio. It should be noted that the larger the value of the security factor ratio, the lower the error rate of the real-time password input, the fewer the number of errors, and the lower the risk of theft. The security factor ratio is compared and analyzed with the preset security factor ratio threshold stored internally:

[0063] If the safety factor ratio is less than the preset safety factor ratio threshold, a high-risk signal is generated and sent to the encryption unit. When the encryption unit receives the risk signal and the high-risk signal, it obtains a self-locking signal and sends the self-locking signal to the display unit through the self-checking and verification unit. After receiving the self-locking signal, the display unit immediately displays the text "Login recognition abnormality" and stops password verification at the same time to ensure the security of the data inside the computer and avoid the risk of information loss.

[0064] If the security factor ratio is greater than or equal to the preset security factor ratio threshold, a pop-up signal is generated and sent to the display unit via the self-checking verification unit. After receiving the pop-up signal, the display unit verifies the computer password again, which helps to ensure the security of password verification.

[0065] Example 4:

[0066] The computer remote login recognition method based on artificial intelligence includes the following steps:

[0067] Step 1: Verify the user ID number and real-time user facial feature image of the login user to obtain an unauthorized signal, a feedback signal, and a secondary verification signal, and then perform corresponding operations based on the feedback signal and the secondary verification signal;

[0068] Step 2: If an unauthorized signal is received, the login operation is terminated. If a feedback signal is received, the operation is carried out in step 3. If a secondary verification signal is received, the real-time password input is verified. If a safe signal is received, the text "Authorized" is displayed. If a risk signal is received, the operation is carried out in step 4.

[0069] Step 3: Collect and analyze operating data to ensure the integrity and validity of the data collected by the recognition camera. If an unqualified signal is received, it will be displayed as "Unauthorized Operation". If a management signal is received, it will be displayed as "Recognition Camera Management". At the same time, the real-time user facial feature image after management will be recognized again.

[0070] Step 4: Perform security risk assessment and analysis on the real-time password input by the login user to determine whether the computer has been stolen and whether the number of verifications is normal. If an input signal is obtained, proceed to step 2. If a comparison instruction is obtained, determine the security risk of the real-time password input. If a high-risk signal is obtained, it will be displayed in the form of text "Login recognition abnormality" and password verification will be stopped at the same time. If a pop-up signal is obtained, proceed to step 2 again.

[0071] In summary, the present invention collects user verification data and analyzes the verification data in an in-depth and feedback manner to improve the security of computational login identification and reduce the risk of information loss. At the same time, the analysis scope is improved by maintaining the real-time user facial feature image, real-time input password and user ID number, which helps to improve the accuracy and comprehensiveness of the analysis results. At the same time, a security risk assessment analysis is performed on the real-time input password of the login through data analysis and feedback to determine whether the computer has been stolen, so as to improve the security supervision effect of the computer, and at the same time improve the rigor of identification verification to ensure the security of the computer's internal data and avoid the risk of information loss. When analyzing the real-time user facial feature image, the operating data of the identification camera is collected and compared in a hierarchical manner to ensure the data collection integrity and effectiveness of the identification camera, which helps to improve the login identification efficiency and improve the supervision and early warning performance of the identification camera.

[0072] The threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technicians in this field for each set of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0073] The above formulas are obtained by collecting a large amount of data and performing software simulation, and a formula close to the actual value is selected. The coefficients in the formula are set by those skilled in the art according to actual conditions. The above is only a preferred specific implementation method of the present invention, but the protection scope of the present invention is not limited to this. Any technician familiar with this technical field, within the technical scope disclosed by the present invention, can make equivalent replacements or changes based on the technical solution and inventive concept of the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A computer remote login recognition system based on artificial intelligence, characterized in that: The system comprises a login platform, which is internally provided with a server, a self-checking and verifying unit, an operation evaluation unit, an encryption unit, a security analysis unit, a database and a display unit; When the server in the login platform generates a login instruction and sends the login instruction to the self-test verification unit, the self-test verification unit immediately collects the user's verification data after receiving the login instruction. The verification data includes a real-time user facial feature image, a real-time input password, and the user's ID number, and analyzes the verification data. The unauthorized signal and the security signal obtained are sent to the display unit, the feedback signal is sent to the operation evaluation unit, and the risk signal obtained is sent to the encryption unit and the security analysis unit; After receiving the feedback signal, the operation evaluation unit immediately collects the operation data of the recognition camera, including the ambient humidity value of the recognition camera, the dust volume on the mirror surface, and the line operating current, analyzes the operation data, sends the obtained unqualified signal to the display unit, and sends the obtained management signal to the display unit and the self-test verification unit; After receiving the risk signal, the security analysis unit immediately obtains the real-time password input by the registrant within the time threshold, performs a security risk assessment analysis on the real-time password input by the registrant, obtains a comparison instruction and input signal, and sends the obtained input signal to the self-checking and verification unit; When the encryption unit receives the risk signal and the high-risk signal, it obtains a self-locking signal and sends the self-locking signal to the display unit through the self-checking and verification unit. After receiving the self-locking signal, the display unit immediately displays the text "Login recognition abnormality" and stops password verification at the same time. The verification data analysis process of the self-test verification unit is as follows: Step 1: The time interval after the server sends the login command is collected and marked as the time threshold. The user ID number and real-time user facial feature image of the login user within the time threshold are obtained. At the same time, the facial feature image corresponding to the login user's user ID number is retrieved from the database and marked as the verification facial feature image; Step 2: Compare and analyze the user ID number and real-time user facial feature image with the internally stored preset user ID number list and verification facial feature image: If the user ID number is not in the preset user ID number list and the real-time user facial feature image is not equal to the verification facial feature image, or if the user ID number is not in the preset user ID number list and the real-time user facial feature image is equal to the verification facial feature image, an unauthorized signal is generated; If the user ID number is within the preset user ID number list and the real-time user facial feature image is not equal to the verification facial feature image, a feedback signal is generated; If the user ID number is within the preset user ID number list and the real-time user facial feature image is equal to the verification facial feature image, a secondary verification signal is generated; When the secondary verification signal is generated in the self-check verification unit, the real-time password input by the login user within the time threshold is immediately obtained, and the real-time password input is compared and analyzed with the preset password stored in the internal input: If the real-time input password is equal to the preset password, a security signal is generated; If the real-time password is not equal to the preset password, a risk signal is generated; The operation data analysis process of the operation evaluation unit is as follows: SS1: Obtain the ambient humidity value HS for the camera within the time threshold, and the dust volume JF on the mirror of the camera within the time threshold, and calculate the interference recognition value GS according to the formula; SS12: Divide the time threshold into i sub-time nodes, where i is a natural number greater than zero. Obtain the line operating current of the identification camera in each sub-time node, construct a set A of line operating currents, and obtain the discrete values ​​of set A. These values ​​are marked as discrete risk values. The maximum and minimum subsets of set A are also obtained, and the difference between the maximum and minimum subsets is marked as the maximum span value. The product of the discrete risk value and the maximum span value is marked as the fault risk value GF. SS13: Obtain the operating status coefficient H according to the formula, and compare and analyze the operating status coefficient H with the preset operating status coefficient threshold stored internally: If the operating state coefficient H is less than or equal to the preset operating state coefficient threshold, a failure signal is generated; If the operating state coefficient H is greater than the preset operating state coefficient threshold, a management signal is generated; The safety risk assessment analysis process of the safety analysis unit is as follows: The real-time password input by the user within the time threshold is obtained, and the digital, alphabetic, and symbol features of the real-time password are extracted. The string of extracted numbers, letters, and symbols is marked as the verification feature character. The total number of verification feature characters is obtained and marked as the verification number. The verification number is compared and analyzed with the preset verification number threshold: If the verification number is not equal to the preset verification number threshold, an input signal is generated; If the verification number is equal to the preset verification number threshold, a comparison instruction is generated; When the security analysis unit generates a comparison instruction, it compares the verification feature character with the preset verification feature character stored in the internal storage one by one: The number of numbers, letters, and symbols that are identical to the preset verification feature characters is obtained and marked as the correct number. The number of numbers, letters, and symbols that are different from the preset verification feature characters is obtained and marked as the incorrect number. The ratio of the correct number to the incorrect number is obtained and marked as the safety factor ratio. The safety factor ratio is compared and analyzed with the preset safety factor ratio threshold value stored internally: If the safety factor ratio is less than the preset safety factor ratio threshold, a high risk signal is generated; If the safety factor ratio is greater than or equal to the preset safety factor ratio threshold, a pop-up signal is generated.

2. A computer remote login recognition method based on artificial intelligence, which is applied to the computer remote login recognition system based on artificial intelligence according to claim 1, characterized in that: The following steps are involved: Step 1: Verify the user ID number and real-time user facial feature image of the login user to obtain an unauthorized signal, a feedback signal, and a secondary verification signal, and then perform corresponding operations based on the feedback signal and the secondary verification signal; Step 2: If an unauthorized signal is received, the login operation is terminated. If a feedback signal is received, the operation is carried out in step 3. If a secondary verification signal is received, the real-time password input is verified. If a safe signal is received, the text "Authorized" is displayed. If a risk signal is received, the operation is carried out in step 4. Step 3: Collect and analyze operational data to ensure the integrity and validity of the camera's data collection. If a failure signal is received, the text "Unauthorized Operation" will be displayed. If a management signal is received, the text "Identification Camera Management" will be displayed. At the same time, the real-time user facial feature image after management will be recognized again. Step 4: Perform a security risk assessment and analysis on the real-time password input by the login user to determine whether the computer has been stolen and whether the number of verifications is normal. If an input signal is obtained, proceed to step 2. If a comparison instruction is obtained, determine the security risk of the real-time password input. If a high-risk signal is obtained, it will be displayed in the form of text "Login recognition abnormality" and password verification will be stopped. If a pop-up signal is obtained, proceed to step 2 again.

Images