A big data-based face recognition screening system and method

Abstract

The present application belongs to the technical field of face recognition, and specifically discloses a face recognition screening system and method based on big data; the embodiment of the present application marks multiple monitoring video data as multiple identification screening monitoring data and multiple safety screening monitoring data; performs access anomaly screening, and when there is access anomaly, performs access anomaly alarm; performs safety behavior monitoring, and when there is abnormal behavior, performs safety anomaly alarm; the multiple monitoring video data can be marked as the multiple identification screening monitoring data and the multiple safety screening monitoring data, the access anomaly screening and the access anomaly alarm are performed through face recognition on the identification screening monitoring data, the safety behavior monitoring and the safety anomaly alarm are performed through safety screening monitoring data, the screening of access anomaly and safety behavior anomaly is automatically performed, the safety anomaly problem in the production of petroleum and chemical industry is discovered in time, and the safety management personnel can process in time.

Description

Technical Field

[0001] This invention belongs to the field of facial recognition technology, and in particular relates to a facial recognition screening system and method based on big data. Background Technology

[0002] Facial recognition is a biometric technology that identifies individuals based on their facial features. It involves using cameras or webcams to capture images or video streams containing faces, automatically detecting and tracking faces within the images, and then performing facial recognition. This process is also commonly referred to as image recognition or face identification. Facial recognition technology integrates various professional technologies such as artificial intelligence, machine recognition, machine learning, model theory, expert systems, and video image processing. It also requires the integration of intermediate value processing theory and implementation. As the latest application of biometric identification, the realization of its core technologies demonstrates the transformation from weak artificial intelligence to strong artificial intelligence.

[0003] The current application of facial recognition technology in the petrochemical industry typically involves recording employees' entry and exit times by recognizing their faces, and using surveillance cameras installed in the petrochemical production area for production monitoring. Safety management and inspection are carried out manually through production monitoring, and alarms are issued for production areas that do not comply with safety regulations. However, this manual safety management and inspection work often fails to detect unsafe behaviors in a timely manner. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a face recognition screening system and method based on big data, which aims to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] A facial recognition screening method based on big data, the method specifically includes the following steps:

[0007] Acquire multiple surveillance video data, perform security identification on the multiple surveillance video data, generate surveillance identification results, and mark the multiple surveillance video data as multiple identification screening surveillance data and multiple security screening surveillance data according to the surveillance identification results;

[0008] Obtain the monitoring location corresponding to the identification and screening monitoring data, and obtain the corresponding identification and screening setting data based on the monitoring location;

[0009] Based on the identification and screening monitoring data, facial recognition is performed to identify multiple on-site staff. Based on the identification and screening settings data, access anomaly screening is performed on the multiple on-site staff, and an access anomaly alarm is triggered when an access anomaly is detected.

[0010] Multiple security identifiers are extracted from the security screening and monitoring data. Based on big data technology, the multiple security identifiers are analyzed to generate security monitoring data.

[0011] Based on the security screening and monitoring data and the security monitoring data, security behavior is monitored to determine whether there is any abnormal behavior, and a security anomaly alarm is triggered when abnormal behavior is detected.

[0012] The following are further optimizations of the above technical solution by the present invention:

[0013] The process of acquiring multiple surveillance video data, performing security identification on the multiple surveillance video data, generating surveillance identification results, and marking the multiple surveillance video data as multiple identification screening surveillance data and multiple security screening surveillance data according to the surveillance identification results specifically includes the following steps:

[0014] Acquire data from multiple surveillance videos;

[0015] Security identifiers are identified from multiple surveillance video data sets to generate surveillance identification results.

[0016] Based on the monitoring and identification results, monitoring video data without security identifiers will be marked as identification and screening monitoring data;

[0017] Based on the monitoring and identification results, monitoring video data with security identifiers are marked as security screening monitoring data.

[0018] Further optimization: The step of obtaining the monitoring location corresponding to the identification and screening monitoring data, and obtaining the corresponding identification and screening setting data based on the monitoring location, specifically includes the following steps:

[0019] Obtain the monitoring sequence number corresponding to the identification and screening monitoring data;

[0020] Based on the monitoring sequence number, determine the monitoring location corresponding to the identification and screening monitoring data;

[0021] Based on the monitored location, match the corresponding identification and screening settings data.

[0022] Further optimization: The step of performing facial recognition based on the identification and screening monitoring data to identify multiple on-site staff, performing access anomaly screening on these staff members based on the identification and screening settings data, and issuing an access anomaly alarm when an access anomaly is detected specifically includes the following steps:

[0023] Based on the identified and screened monitoring data, facial recognition was performed to identify multiple on-site staff members;

[0024] Based on the identification and screening settings data, an access whitelist is determined;

[0025] Based on the access whitelist, multiple on-site staff were screened for access anomalies to determine whether any access anomalies existed.

[0026] An access anomaly alarm will be triggered when an access anomaly is detected.

[0027] Further optimization: The extraction of multiple security identifiers from the security screening and monitoring data, and the analysis of these multiple security identifiers based on big data technology to generate security monitoring data specifically includes the following steps:

[0028] Extract the security screening monitoring images from the security screening monitoring data;

[0029] Extract multiple security icons from the security screening and monitoring images;

[0030] Based on big data technology, multiple security identifiers are analyzed to generate security monitoring data.

[0031] Further optimization: The step of monitoring security behavior based on the security screening and monitoring data and the security monitoring data, determining whether abnormal behavior exists, and issuing a security anomaly alarm when abnormal behavior is found specifically includes the following steps:

[0032] The aforementioned safety screening and monitoring data are analyzed to obtain real-time operational data;

[0033] Based on the security monitoring data, security behavior monitoring is performed on the real-time working data to determine whether any abnormal behavior exists;

[0034] When abnormal behavior is detected, a security anomaly alarm will be triggered.

[0035] This invention also provides a face recognition screening system based on big data. The system includes a monitoring video marking unit, a data acquisition unit, an access anomaly screening unit, a security identifier analysis unit, and a security behavior monitoring unit, wherein:

[0036] A surveillance video tagging unit is used to acquire multiple surveillance video data, perform security identification on the multiple surveillance video data, generate surveillance identification results, and tag the multiple surveillance video data as multiple identification screening surveillance data and multiple security screening surveillance data according to the surveillance identification results.

[0037] A data acquisition unit is set up to acquire the monitoring location corresponding to the identification and screening monitoring data, and to acquire the corresponding identification and screening setting data according to the monitoring location.

[0038] The access anomaly screening unit is used to perform facial recognition based on the identification and screening monitoring data, identify multiple on-site staff, perform access anomaly screening on the multiple on-site staff based on the identification and screening setting data, and issue an access anomaly alarm when an access anomaly is found.

[0039] The safety identifier analysis unit is used to extract multiple safety identifiers from the safety screening and monitoring data, analyze the multiple safety identifiers based on big data technology, and generate safety monitoring data.

[0040] The safety behavior monitoring unit is used to monitor safety behavior based on the safety screening monitoring data and the safety monitoring data, determine whether there is any abnormal behavior, and issue a safety anomaly alarm when abnormal behavior is found.

[0041] The following are further optimizations of the above technical solution by the present invention:

[0042] The surveillance video marking unit specifically includes:

[0043] The monitoring data acquisition module is used to acquire multiple monitoring video data.

[0044] The security identifier recognition module is used to identify security identifiers in multiple surveillance video data sets and generate surveillance identification results.

[0045] The first data marking module is used to mark surveillance video data without security identifiers as identification and screening surveillance data based on the monitoring identification results;

[0046] The second data module is used to mark surveillance video data with security identifiers as security screening surveillance data based on the surveillance identification results.

[0047] Further optimization: The access anomaly screening unit specifically includes:

[0048] The monitoring face recognition module is used to perform face recognition based on the recognition and screening monitoring data to identify multiple on-site staff.

[0049] The whitelist determination module is used to determine the access whitelist based on the identification and screening settings data;

[0050] The access anomaly screening module is used to screen multiple on-site staff for access anomalies based on the access whitelist and determine whether there are any access anomalies.

[0051] The access anomaly alarm module is used to issue an access anomaly alarm when an access anomaly is detected.

[0052] Further optimization: The safety behavior monitoring unit specifically includes:

[0053] The work data acquisition module is used to analyze the safety screening and monitoring data to obtain real-time work data;

[0054] The safety behavior monitoring module is used to monitor the real-time working data for safety behavior based on the safety monitoring data, and to determine whether there is any abnormal behavior.

[0055] The safety anomaly alarm module is used to issue a safety anomaly alarm when abnormal behavior is detected.

[0056] The present invention, by adopting the above technical solution, has the following beneficial effects:

[0057] This invention, in its embodiments, marks multiple surveillance video data sets as multiple identification and screening monitoring data sets and multiple safety screening monitoring data sets, respectively; performs access anomaly screening and issues an access anomaly alarm when an access anomaly is detected; and performs safety behavior monitoring and issues a safety anomaly alarm when abnormal behavior is detected. It can automatically screen for access anomalies and safety behavior anomalies by marking multiple surveillance video data sets as multiple identification and screening monitoring data sets and multiple safety screening monitoring data sets, and by performing access anomaly screening and alarm on the identification and screening monitoring data sets, and by performing safety behavior monitoring and alarm on the safety screening monitoring data sets. This enables timely detection of safety anomalies in petrochemical production, facilitating timely handling by safety management personnel.

[0058] The present invention will be further described below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0059] Figure 1 A flowchart of the method provided by an embodiment of the present invention is shown.

[0060] Figure 2 A flowchart illustrating the classification and labeling of surveillance video data in the method provided by an embodiment of the present invention is shown.

[0061] Figure 3 A flowchart illustrating the method for obtaining identification screening settings data provided in an embodiment of the present invention is shown.

[0062] Figure 4 A flowchart of the access anomaly screening method provided in the embodiments of the present invention is shown.

[0063] Figure 5 A flowchart of the security identifier analysis in the method provided by an embodiment of the present invention is shown.

[0064] Figure 6 A flowchart of the safety behavior monitoring method provided in the embodiments of the present invention is shown.

[0065] Figure 7An application architecture diagram of the system provided in an embodiment of the present invention is shown.

[0066] Figure 8 A structural block diagram of the monitoring video marking unit in the system provided by an embodiment of the present invention is shown.

[0067] Figure 9 The diagram shows a structural block diagram of the access anomaly screening unit in the system provided by an embodiment of the present invention.

[0068] Figure 10 A structural block diagram of the safety behavior monitoring unit in the system provided by an embodiment of the present invention is shown. Detailed Implementation

[0069] like Figure 1 As shown, a face recognition screening method based on big data includes the following steps:

[0070] Step S101: Acquire multiple surveillance video data, perform security identification on the multiple surveillance video data, generate surveillance identification results, and mark the multiple surveillance video data as multiple identification screening surveillance data and multiple security screening surveillance data according to the surveillance identification results.

[0071] In this embodiment of the invention, video monitoring points are set up in multiple locations in the petrochemical plant, and real-time monitoring and shooting are carried out at multiple video monitoring points to acquire multiple monitoring video data. Safety identification is performed on the multiple monitoring video data to generate monitoring identification results. Based on the monitoring identification results, it is determined whether there are safety identifications in the monitoring video data. Monitoring video data with safety identifications are marked as safety screening monitoring data, and monitoring video data without safety identifications are marked as identification screening monitoring data.

[0072] It is understandable that there are multiple video surveillance points in a petrochemical plant, including area identification surveillance points and safety production surveillance points. The safety production surveillance points have different safety signs posted according to specific production safety requirements. Therefore, by judging whether there are safety signs in the surveillance video data, multiple surveillance video data can be classified and marked as multiple identification screening surveillance data and multiple safety screening surveillance data.

[0073] like Figure 2 As shown, the steps of acquiring multiple surveillance video data, performing security identifier recognition on the multiple surveillance video data, generating surveillance identification results, and marking the multiple surveillance video data as multiple identification screening surveillance data and multiple security screening surveillance data according to the surveillance identification results specifically include the following steps:

[0074] Step S1011: Obtain multiple surveillance video data.

[0075] Step S1012: Perform security identification on multiple monitoring video data to generate monitoring identification results.

[0076] Step S1013: Based on the monitoring identification results, the monitoring video data without security identifiers are marked as identification screening monitoring data.

[0077] Step S1014: Based on the monitoring and identification results, the monitoring video data with security identifiers are marked as security screening monitoring data.

[0078] The big data-based facial recognition screening method also includes the following steps:

[0079] Step S102: Obtain the monitoring location corresponding to the identification and screening monitoring data, and obtain the corresponding identification and screening setting data according to the monitoring location.

[0080] In this embodiment of the invention, when acquiring identification screening monitoring data, the identification screening monitoring data is analyzed to determine the monitoring sequence number corresponding to the identification screening monitoring data. According to the monitoring sequence number, the monitoring position of the area entry and exit monitoring point corresponding to the identification screening monitoring data is determined, and then the corresponding identification screening setting data is matched according to the monitoring position.

[0081] It is understandable that the identification and screening settings data are preset and are used to compare the monitoring and screening data of relevant monitoring locations. The identification and screening settings data contain an access whitelist, and only people on the access whitelist can enter the work area corresponding to the relevant monitoring location.

[0082] like Figure 3 As shown, obtaining the monitoring location corresponding to the identification and screening monitoring data, and obtaining the corresponding identification and screening setting data based on the monitoring location, specifically includes the following steps:

[0083] Step S1021: Obtain the monitoring sequence number corresponding to the identification screening monitoring data.

[0084] Step S1022: Determine the monitoring location corresponding to the identification and screening monitoring data based on the monitoring sequence number.

[0085] Step S1023: Match the corresponding identification and screening settings data according to the monitoring location.

[0086] The big data-based facial recognition screening method also includes the following steps:

[0087] Step S103: Perform facial recognition based on the identification and screening monitoring data to identify multiple on-site staff members, perform access anomaly screening on the multiple on-site staff members based on the identification and screening settings data, and issue an access anomaly alarm when an access anomaly is detected.

[0088] In this embodiment of the invention, facial recognition is performed on personnel at relevant monitoring locations based on identification and screening monitoring data to identify multiple on-site staff. Based on the identification and screening settings data, an access whitelist for the relevant monitoring locations is determined. By comparing the multiple on-site staff identified by facial recognition with the access whitelist, it is determined whether the multiple on-site staff meet the requirements of the access whitelist. If there are on-site staff who do not meet the requirements of the access whitelist, an access anomaly alarm is triggered at the relevant monitoring location.

[0089] It is understandable that the monitoring location corresponding to the identification and screening monitoring data is set at a certain location within the relevant area, and is located between the entrance / exit and the work location, which can realize secondary monitoring, identification and screening of the relevant area in addition to the verification of entrance / exit.

[0090] Understandably, if an employee deliberately covers their face to avoid facial recognition, the employee who cannot be recognized will be identified as someone who does not meet the access whitelist requirements.

[0091] like Figure 4 As shown, the process of performing facial recognition based on the identification and screening monitoring data to identify multiple on-site staff, conducting access anomaly screening on these staff members based on the identification and screening settings data, and issuing an access anomaly alarm when an access anomaly is detected specifically includes the following steps:

[0092] Step S1031: Perform facial recognition based on the identified screening and monitoring data to identify multiple on-site staff.

[0093] Step S1032: Determine the access whitelist based on the identification and screening settings data.

[0094] Step S1033: Based on the access whitelist, conduct access anomaly screening on multiple on-site staff to determine whether there are any access anomalies.

[0095] Step S1034: If an access anomaly exists, trigger an access anomaly alarm.

[0096] The big data-based facial recognition screening method also includes the following steps:

[0097] Step S104: Extract multiple security identifiers from the security screening and monitoring data, and analyze the multiple security identifiers based on big data technology to generate security monitoring data.

[0098] In this embodiment of the invention, the monitoring video corresponding to the security screening monitoring data is processed frame by frame to extract the clearest security screening monitoring image of the security sign area. Then, multiple security signs are extracted from the security screening monitoring image. Based on big data technology, the multiple security signs are analyzed to generate security monitoring data.

[0099] It is understandable that safety signs are general text and / or images. Big data technology can be used to analyze safety signs to identify prohibited behaviors that do not meet the safety requirements of the corresponding area, and then generate safety monitoring data to monitor the relevant prohibited behaviors.

[0100] like Figure 5 As shown, the steps for extracting multiple security identifiers from the security screening and monitoring data, analyzing these identifiers based on big data technology, and generating security monitoring data specifically include the following:

[0101] Step S1041: Capture the security screening monitoring image from the security screening monitoring data.

[0102] Step S1042: Extract multiple security icons from the security screening monitoring image.

[0103] Step S1043: Based on big data technology, analyze the multiple security identifiers to generate security monitoring data.

[0104] The big data-based facial recognition screening method also includes the following steps:

[0105] Step S105: Based on the security screening monitoring data and the security monitoring data, perform security behavior monitoring to determine whether there is any abnormal behavior, and issue a security anomaly alarm when abnormal behavior is found.

[0106] In this embodiment of the invention, by analyzing the safety screening and monitoring data, real-time work data reflecting the work behavior of staff is obtained. Based on the safety monitoring data, anomaly analysis is performed on the real-time work data to determine whether there is any abnormal behavior, and a safety anomaly alarm is triggered when abnormal behavior is detected.

[0107] It is understandable that abnormal behavior is the prohibited behavior in the corresponding area.

[0108] like Figure 6 As shown, the step of monitoring security behavior based on the security screening and monitoring data and the security monitoring data, determining whether there is abnormal behavior, and issuing a security anomaly alarm when abnormal behavior is found specifically includes the following steps:

[0109] Step S1051: Analyze the security screening and monitoring data to obtain real-time working data.

[0110] Step S1052: Based on the security monitoring data, perform security behavior monitoring on the real-time working data to determine whether there is any abnormal behavior.

[0111] Step S1053: When abnormal behavior is detected, a security anomaly alarm is triggered.

[0112] like Figure 7 As shown, the present invention also provides a face recognition screening system based on big data, comprising:

[0113] The monitoring video tagging unit 101 is used to acquire multiple monitoring video data, perform security identification on the multiple monitoring video data, generate monitoring identification results, and tag the multiple monitoring video data as multiple identification screening monitoring data and multiple security screening monitoring data according to the monitoring identification results.

[0114] In this embodiment of the invention, video monitoring points are set up in multiple locations in the petrochemical plant, and real-time monitoring and shooting are carried out at multiple video monitoring points. The monitoring video marking unit 101 acquires multiple monitoring video data, performs safety mark identification on the multiple monitoring video data, generates monitoring identification results, and determines whether there are safety marks in the monitoring video data based on the monitoring identification results. The monitoring video data with safety marks is marked as safety screening monitoring data, and the monitoring video data without safety marks is marked as identification screening monitoring data.

[0115] like Figure 8 As shown, the surveillance video marking unit 101 specifically includes:

[0116] The monitoring data acquisition module 1011 is used to acquire multiple monitoring video data.

[0117] The security identification module 1012 is used to identify security identifications in multiple surveillance video data and generate surveillance identification results.

[0118] The first data marking module 1013 is used to mark surveillance video data without security identifiers as identification screening surveillance data based on the monitoring identification results.

[0119] The second data module 1014 is used to mark surveillance video data with security identifiers as security screening surveillance data based on the surveillance identification results.

[0120] The big data-based facial recognition screening system also includes:

[0121] The data acquisition unit 102 is configured to acquire the monitoring location corresponding to the identification and screening monitoring data, and acquire the corresponding identification and screening setting data based on the monitoring location.

[0122] In this embodiment of the invention, when acquiring identification screening monitoring data, the data acquisition unit 102 analyzes the identification screening monitoring data, determines the monitoring sequence number corresponding to the identification screening monitoring data, determines the monitoring position of the area entry and exit monitoring point corresponding to the identification screening monitoring data according to the monitoring sequence number, and then matches the corresponding identification screening setting data according to the monitoring position.

[0123] The access anomaly screening unit 103 is used to perform facial recognition based on the identification and screening monitoring data, identify multiple on-site staff, perform access anomaly screening on the multiple on-site staff based on the identification and screening setting data, and issue an access anomaly alarm when an access anomaly is found.

[0124] In this embodiment of the invention, the access anomaly screening unit 103 performs facial recognition on personnel at relevant monitoring locations based on the identification and screening monitoring data to identify multiple on-site staff. Based on the identification and screening settings data, it determines the access whitelist for the relevant monitoring locations. By comparing the multiple on-site staff identified by facial recognition with the access whitelist, it determines whether the multiple on-site staff meet the requirements of the access whitelist. If there are on-site staff who do not meet the requirements of the access whitelist, an access anomaly alarm is triggered at the relevant monitoring location.

[0125] like Figure 9 As shown, the access anomaly screening unit 103 specifically includes:

[0126] The monitoring face recognition module 1031 is used to perform face recognition based on the recognition screening monitoring data to identify multiple on-site staff.

[0127] The whitelist determination module 1032 is used to determine the access whitelist based on the identification and screening settings data.

[0128] The access anomaly screening module 1033 is used to screen multiple on-site staff for access anomalies based on the access whitelist and determine whether there are any access anomalies.

[0129] The access anomaly alarm module 1034 is used to issue an access anomaly alarm when an access anomaly exists.

[0130] The big data-based facial recognition screening system also includes:

[0131] The safety identifier analysis unit 104 is used to extract multiple safety identifiers from the safety screening and monitoring data, analyze the multiple safety identifiers based on big data technology, and generate safety monitoring data.

[0132] In this embodiment of the invention, the safety identification analysis unit 104 performs frame-by-frame processing on the monitoring video corresponding to the safety screening monitoring data, extracts the clearest safety screening monitoring image of the safety identification area, and then extracts multiple safety identifications from the safety screening monitoring image. Based on big data technology, it analyzes the multiple safety identifications to generate safety monitoring data.

[0133] The safety behavior monitoring unit 105 is used to monitor safety behavior based on the safety screening monitoring data and the safety monitoring data, determine whether there is any abnormal behavior, and issue a safety anomaly alarm when abnormal behavior is found.

[0134] In this embodiment of the invention, the safety behavior monitoring unit 105 analyzes the safety screening and monitoring data to obtain real-time work data reflecting the work behavior of the staff, performs anomaly analysis on the real-time work data according to the safety monitoring data, determines whether there is any abnormal behavior, and issues a safety anomaly alarm when there is abnormal behavior.

[0135] Figure 10 As shown, the safety behavior monitoring unit 105 specifically includes:

[0136] The work data acquisition module 1051 is used to analyze the safety screening and monitoring data to obtain real-time work data.

[0137] The safety behavior monitoring module 1052 is used to monitor the real-time working data for safety behavior based on the safety monitoring data, and to determine whether there is any abnormal behavior.

[0138] The safety anomaly alarm module 1053 is used to issue a safety anomaly alarm when abnormal behavior occurs.

[0139] In summary, the embodiments of the present invention can mark multiple monitoring video data as multiple identification screening monitoring data and multiple safety screening monitoring data respectively. By performing facial recognition on the identification screening monitoring data for access anomaly screening and access anomaly alarm, and by performing safety behavior monitoring and safety anomaly alarm on the safety screening monitoring data, the invention can automatically screen for access anomalies and safety behavior anomalies, promptly detect safety anomalies in petrochemical production, and facilitate timely handling by safety management personnel.

[0140] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0141] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0142] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0143] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0144] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A face recognition screening method based on big data, characterized in that: The method specifically includes the following steps: Acquire multiple surveillance video data, perform security identification on the multiple surveillance video data, generate surveillance identification results, and mark the multiple surveillance video data as multiple identification screening surveillance data and multiple security screening surveillance data according to the surveillance identification results; Obtain the monitoring location corresponding to the identification and screening monitoring data, and obtain the corresponding identification and screening setting data based on the monitoring location; Based on the identification and screening monitoring data, facial recognition is performed to identify multiple on-site staff. Based on the identification and screening settings data, access anomaly screening is performed on the multiple on-site staff, and an access anomaly alarm is triggered when an access anomaly is detected. Multiple security identifiers are extracted from the security screening and monitoring data. Based on big data technology, the multiple security identifiers are analyzed to generate security monitoring data. Based on the security screening and monitoring data and the security monitoring data, security behavior is monitored to determine whether there is any abnormal behavior, and a security anomaly alarm is triggered when abnormal behavior is detected. The process of acquiring multiple surveillance video data, performing security identification on the multiple surveillance video data, generating surveillance identification results, and marking the multiple surveillance video data as multiple identification screening surveillance data and multiple security screening surveillance data according to the surveillance identification results specifically includes the following steps: Acquire data from multiple surveillance videos; Security identifiers are identified from multiple surveillance video data sets to generate surveillance identification results. Based on the monitoring and identification results, monitoring video data without security identifiers will be marked as identification and screening monitoring data; Based on the monitoring and identification results, monitoring video data with security identifiers will be marked as security screening monitoring data; The step of obtaining the monitoring location corresponding to the identification and screening monitoring data, and obtaining the corresponding identification and screening setting data based on the monitoring location, specifically includes the following steps: Obtain the monitoring sequence number corresponding to the identification and screening monitoring data; Based on the monitoring sequence number, determine the monitoring location corresponding to the identification and screening monitoring data; Based on the monitored location, match the corresponding identification and screening settings data.

2. The face recognition screening method based on big data according to claim 1, characterized in that: The process of performing facial recognition based on the identification and screening monitoring data to identify multiple on-site staff, screening for access anomalies among these staff members based on the identification and screening settings data, and issuing an access anomaly alarm when an access anomaly is detected specifically includes the following steps: Based on the identified and screened monitoring data, facial recognition was performed to identify multiple on-site staff members; Based on the identification and screening settings data, an access whitelist is determined; Based on the access whitelist, multiple on-site staff were screened for access anomalies to determine whether any access anomalies existed. An access anomaly alarm will be triggered when an access anomaly is detected.

3. The face recognition screening method based on big data according to claim 1, characterized in that: The process of extracting multiple security identifiers from the security screening and monitoring data, and analyzing these identifiers based on big data technology to generate security monitoring data specifically includes the following steps: Extract the security screening monitoring images from the security screening monitoring data; Extract multiple security icons from the security screening and monitoring images; Based on big data technology, multiple security identifiers are analyzed to generate security monitoring data.

4. The face recognition screening method based on big data according to claim 1, characterized in that: The step of monitoring safety behavior based on the safety screening and monitoring data and the safety monitoring data, determining whether there is any abnormal behavior, and issuing a safety anomaly alarm when abnormal behavior is found specifically includes the following steps: The security screening and monitoring data are analyzed to obtain real-time operational data; Based on the security monitoring data, security behavior monitoring is performed on the real-time working data to determine whether any abnormal behavior exists; When abnormal behavior is detected, a security anomaly alarm will be triggered.

5. A facial recognition screening system based on big data, characterized in that: The system includes a video monitoring and tagging unit, a data acquisition unit, an access anomaly screening unit, a security identification analysis unit, and a security behavior monitoring unit, wherein: A surveillance video tagging unit is used to acquire multiple surveillance video data, perform security identification on the multiple surveillance video data, generate surveillance identification results, and tag the multiple surveillance video data as multiple identification screening surveillance data and multiple security screening surveillance data according to the surveillance identification results. A data acquisition unit is set up to acquire the monitoring location corresponding to the identification and screening monitoring data, and to acquire the corresponding identification and screening setting data according to the monitoring location. The access anomaly screening unit is used to perform facial recognition based on the identification and screening monitoring data, identify multiple on-site staff, perform access anomaly screening on the multiple on-site staff based on the identification and screening setting data, and issue an access anomaly alarm when an access anomaly is found. The safety identifier analysis unit is used to extract multiple safety identifiers from the safety screening and monitoring data, analyze the multiple safety identifiers based on big data technology, and generate safety monitoring data. The safety behavior monitoring unit is used to monitor safety behavior based on the safety screening monitoring data and the safety monitoring data, determine whether there is any abnormal behavior, and issue a safety anomaly alarm when abnormal behavior is found. The surveillance video marking unit specifically includes: The monitoring data acquisition module is used to acquire multiple monitoring video data. The security identifier recognition module is used to identify security identifiers in multiple surveillance video data sets and generate surveillance identification results. The first data marking module is used to mark surveillance video data without security identifiers as identification and screening surveillance data based on the monitoring identification results; The second data module is used to mark surveillance video data with security identifiers as security screening surveillance data based on the surveillance identification results.

6. The face recognition screening system based on big data according to claim 5, characterized in that: The access anomaly screening unit specifically includes: The monitoring face recognition module is used to perform face recognition based on the recognition and screening monitoring data to identify multiple on-site staff. The whitelist determination module is used to determine the access whitelist based on the identification and screening settings data; The access anomaly screening module is used to screen multiple on-site staff for access anomalies based on the access whitelist and determine whether there are any access anomalies. The access anomaly alarm module is used to issue an access anomaly alarm when an access anomaly is detected.

7. The face recognition screening system based on big data according to claim 5, characterized in that: The safety behavior monitoring unit specifically includes: The work data acquisition module is used to analyze the safety screening and monitoring data to acquire real-time work data; The safety behavior monitoring module is used to monitor the real-time working data for safety behavior based on the safety monitoring data, and to determine whether there is any abnormal behavior. The safety anomaly alarm module is used to issue a safety anomaly alarm when abnormal behavior is detected.

Images