A human posture recognition security system based on edge computing

Abstract

The embodiment of the present invention discloses a human posture recognition security system based on edge computing, which relates to the field of edge computing technology. The system includes: a camera installed on the target access control, used to collect the walking image of the current person; an edge computing system, arranged in the camera, used to detect whether the human posture of the current person in the walking image of the current person is in a normal state, and determine the current person's access authority to the target access control according to the preset recognition algorithm and the human posture detection result of the current person; a door opening control module, connected to the edge computing system, used to control the target access control to open when the edge computing system determines to control the release of the current person's access authority to the target access control, so as to allow the current person to enter. The present invention can effectively enhance the security of the access control system.

Description

Technical Field

[0001] The present invention belongs to the field of edge computing technology, and in particular relates to a human body posture recognition security system based on edge computing. Background Art

[0002] Access control system, as the name implies, is a system that controls the entrance and exit channels to prevent illegal personnel from entering the protected area / place, thereby ensuring the safety of the area / place. With the development of induction card technology and biometric technology, access control system has made a leap forward and entered a mature stage. There are various technical systems such as password keyboard access control system, induction card access control system, fingerprint access control system, finger vein recognition access control system, face recognition access control system, etc., making the application field of access control system more and more extensive.

[0003] However, in terms of security performance, the current access control system cannot meet the needs of scenarios with extremely high security requirements. For example, password keypad access control, induction card access control system, fingerprint access control system, finger vein recognition access control system, etc. can all be cracked by copying the password / biometric features of legitimate users or by brute force, which can easily lead to security risks; face recognition access control system is based on face recognition technology, but face recognition technology has the problem of misrecognition, especially in poor lighting conditions, which will lead to an increase in the misrecognition rate. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a human posture recognition security system based on edge computing, which is used to solve the problem of low security of existing access control systems. The present invention verifies the legitimacy of the current person based on human posture information, and opens the access control after the verification is passed, which can effectively enhance security, expand the application scope of the access control system in some specific scenarios, and expand the diversity of security measures.

[0005] The embodiment of the present invention provides a human posture recognition security system based on edge computing, including:

[0006] The camera is installed on the target access control and is used to collect the current walking image of the person;

[0007] An edge computing system, arranged in the camera, is used to detect whether the body posture of the current person in the walking image of the current person is normal, and determine the access permission of the current person to the target access control according to a preset recognition algorithm and the body posture detection result of the current person;

[0008] The door opening control module is connected to the edge computing system and is used to control the target access control to open when the edge computing system determines to release the current person's access to the target access control so that the current person can enter.

[0009] In an optional embodiment, the human body posture includes at least one of: a walking posture of a person, a size of a person's steps, and an amplitude of a person's arm swing;

[0010] The edge computing system is specifically used to detect whether the human body posture of the current person in the walking image of the current person is normal according to a preset normal human body posture library.

[0011] In an optional embodiment, the human posture recognition security system based on edge computing further includes:

[0012] A record storage module is used to store the detection record of the detected person when the edge computing system determines to control and close the current person's access to the target door; the detection record includes the identification information of the detected person and the human body posture detection result of the detected person.

[0013] In an optional embodiment, the edge computing system includes:

[0014] The posture detection unit is used to identify the human posture of the current person in the walking image of the current person, and detect whether the human posture of the current person is normal, and obtain the detection result;

[0015] A face recognition unit is used to perform face recognition on the walking image of the current person to obtain the face feature information of the current person;

[0016] A first judgment unit, configured to judge whether the facial feature information of the current person matches any recorded person according to the person detection record stored in the storage record module;

[0017] an authority determination unit, for obtaining, when the judgment result of the first judgment unit is yes, a historical detection record of a recorded person that matches the facial feature information of the current person, and determining the access authority of the current person to the target access control system based on the currently obtained historical detection record and the human body posture detection result of the current person;

[0018] The identification information generating unit is used to generate the identification information of the current person according to the facial feature information of the current person when the judgment result of the first judgment unit is no, and send it to the storage recording module for storage as the identification information of the person being detected this time.

[0019] In an optional embodiment, the first judgment unit is specifically used to calculate the matching judgment value between the facial feature information of the current person at the current moment and the recorded person according to the first formula, and judge whether the matching judgment value between the facial feature information of the current person at the current moment and the recorded person is equal to 0, if so, judge that the facial feature information of the current person matches the recorded person, otherwise, judge that the facial feature information of the current person does not match the recorded person;

[0020] Among them, the first formula is:

[0021]

[0022] In the first formula, E(t) represents the matching judgment value between the facial feature information of the current person at the current moment and the recorded person; t represents the current moment; G 16 The hexadecimal form of the facial feature information of the current person output by the facial recognition unit; D 16 (i) represents the hexadecimal form of the identification information of the i-th recorded person stored in the storage and recording module; () 16 Indicates converting the value in the brackets to hexadecimal form; len[] indicates the number of bits of the hexadecimal data in the brackets; << indicates left shift; i = 1, 2, ..., n(t); n(t) indicates the total number of people recorded at the current moment; R{,} indicates judging the facial similarity of the data on both sides of the comma in the brackets. If it is judged to be the same face, the value 1 is output, and if it is judged to be not the same face, the value 0 is output;

[0023] The identification information generating unit is specifically used to generate identification information of a person according to the second formula;

[0024] Wherein, the second formula is:

[0025] D 16 [n(t)+1]=[n(t)+1] 16 < <len(G 16 )+G 16

[0026] In the second formula, D 16 [n(t)+1] indicates that when the facial feature information of the current person does not match any of the n(t) recorded persons, the identification information generating unit generates the identification information of the current person in hexadecimal form, and sends it to the storage and recording module so that it is stored and recorded as the identification information of the n(t)+1th recorded person; [] 16 Indicates converting the value in brackets to hexadecimal format.

[0027] In an optional embodiment, the authority determination unit includes:

[0028] a person determination subunit, configured to determine, when the judgment result of the first judgment unit is yes, a recorded person matching the facial feature information of the current person according to a third formula as a target recorded person;

[0029] A detection record acquisition subunit, used to acquire the historical detection records of the target record personnel stored in the storage record module;

[0030] An entry authority control value calculation subunit, used to calculate the entry authority control value of the current person at the current moment according to the fourth formula;

[0031] The judgment subunit is used to judge whether the access permission control value of the current person at the current moment is equal to 0. If so, it is determined to control closing the access permission of the current person to the target access control. Otherwise, it is determined to control releasing the access permission of the current person to the target access control.

[0032] Wherein, the third formula is:

[0033]

[0034] In the third formula, I indicates that the target recorded person determined at the current moment is the Ith recorded person; It means to substitute the value of i from 1 to n(t) into the brackets to obtain the value of i that makes the brackets true;

[0035] The fourth formula is:

[0036]

[0037] In the fourth formula, Q(t) represents the access permission control value of the current person at the current moment; δ[] represents the 0 judgment function, if the value in the bracket is 0, the function value is 1, if the value in the bracket is not 0, the function value is 0; J I (e) represents the human posture detection result recorded in the e-th historical detection record of the target recorded person. If the human posture of the target recorded person is detected to be normal in the e-th detection, then J I (e) = 0, otherwise J I (e)=1;e=1,2,…,B I ; B I It represents the total number of historical detection records of the target person acquired by the detection record acquisition subunit; J0 represents the body posture detection result of the current person.

[0038] In an optional embodiment, the human posture recognition security system based on edge computing further includes:

[0039] An alarm module is used to send an alarm message to a designated security monitoring terminal when the edge computing system determines to control and close the access rights of the current person to the target access control; wherein the alarm message includes the target access control identification, the walking image of the current person, and the access rights of the current person to the target access control determined by the edge computing system of the target access control;

[0040] The door opening control module is further used to control the target door to open when receiving the user door opening instruction returned by the designated security monitoring terminal according to the alarm information.

[0041] The present invention provides a human posture recognition security system based on edge computing. First, a camera is used to collect a walking image of the current person in front of the access control. Then, through the edge computing system, according to a preset recognition algorithm and the human posture detection result of the current person in the walking image, the current person's access to the access control is determined. Finally, when the current person has access rights, the access control is controlled to be opened. The present invention verifies the legitimacy of the current person based on human posture information, and opens the access control after the verification is passed. It can effectively strengthen security protection, expand the application scope of the access control system in some specific scenarios, and expand the diversity of security measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 A schematic diagram of the structure of a human body posture recognition security system based on edge computing provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0045] It should be clear that the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] Figure 1 A schematic diagram of a human posture recognition security system based on edge computing provided by an embodiment of the present invention. Figure 1 The system comprises:

[0047] Camera 1, installed on the target access control, is used to collect the current person's walking image;

[0048] The edge computing system 2 is arranged in the camera 1, and is used to detect whether the human body posture of the current person in the walking image of the current person is in a normal state, and determine the current person's access to the target access control according to a preset recognition algorithm and the human body posture detection result of the current person; preferably, the human body posture includes: at least one of the walking posture of the person, the size of the person's steps, and the amplitude of the person's arm swing; the edge computing system 2 is specifically used to detect whether the human body posture of the current person in the walking image of the current person is in a normal state according to a preset normal human body posture library;

[0049] The door opening control module 3 is connected to the edge computing system 2, and is used to control the target access control to open when the edge computing system 2 determines to release the current person's access to the target access control, so as to allow the current person to enter.

[0050] The beneficial effects of the above technical solution are as follows: the embodiment of the present invention provides a human posture recognition security system based on edge computing, which first collects the walking image of the current person in front of the access control through camera 1, and then determines the current person's access to the access control through edge computing system 2 according to the preset recognition algorithm and the human posture detection result of the current person in the walking image, and finally controls the access control to be opened when the current person has access authority. The present invention verifies the legitimacy of the current person based on human posture information, and opens the access control after the verification is passed, which can effectively strengthen security protection, expand the application scope of the access control system in some specific scenarios, and expand the diversity of security measures.

[0051] As an optional embodiment, the human posture recognition security system based on edge computing further includes:

[0052] The record storage module is used to store the detection record of the detected person when the edge computing system 2 determines to control and close the current person's access to the target door; the detection record includes the identification information of the detected person and the human body posture detection result of the detected person.

[0053] The beneficial effect of the above technical solution is: based on the body posture information of the current person in front of the target access control, when it is detected that the current person is an illegal person, the target access control is first controlled to be closed, and the identification information of the current person and the body posture detection results are recorded, thereby forming a closed-loop calculation and control to ensure the reliability of the access control system.

[0054] As an optional embodiment, the edge computing system 2 includes:

[0055] The posture detection unit is used to identify the human posture of the current person in the walking image of the current person, and detect whether the human posture of the current person is normal, and obtain the detection result;

[0056] A face recognition unit is used to perform face recognition on the walking image of the current person to obtain the face feature information of the current person;

[0057] A first judgment unit, configured to judge whether the facial feature information of the current person matches any recorded person according to the person detection record stored in the storage record module;

[0058] an authority determination unit, for obtaining, when the judgment result of the first judgment unit is yes, a historical detection record of a recorded person that matches the facial feature information of the current person, and determining the access authority of the current person to the target access control system based on the currently obtained historical detection record and the human body posture detection result of the current person;

[0059] The identification information generating unit is used to generate the identification information of the current person according to the facial feature information of the current person when the judgment result of the first judgment unit is no, and send it to the storage recording module for storage as the identification information of the person being detected this time.

[0060] The beneficial effects of the above technical solution are: using face recognition technology to identify whether the facial feature information of the current person in front of the access control system matches the facial feature information of the recorded person (i.e., the legal user). If so, it is preliminarily determined that the current person may be a legal user, otherwise the current person is directly determined to be an illegal user; further, when the current person may be a legal user, the current person's access authority to the target access control system is determined based on the historical detection records of the recorded persons that match the facial feature information of the current person and the human posture detection results, thereby realizing face recognition and human posture recognition of the current person, which not only effectively improves the performance of the system, but also effectively improves the security of the system.

[0061] As an optional embodiment, the first judgment unit is specifically used to calculate the matching judgment value between the facial feature information of the current person at the current moment and the recorded person according to the first formula, and judge whether the matching judgment value between the facial feature information of the current person at the current moment and the recorded person is equal to 0, if so, judge that the facial feature information of the current person matches the recorded person, otherwise, judge that the facial feature information of the current person does not match the recorded person;

[0062] Among them, the first formula is:

[0063]

[0064] In the first formula, E(t) represents the matching judgment value between the facial feature information of the current person at the current moment and the recorded person. If E(t)=0, it is determined that the facial feature information of the current person matches the recorded person. If E(t)≠0, it is determined that the facial feature information of the current person does not match the recorded person. t represents the current moment. G 16 The hexadecimal form of the facial feature information of the current person output by the facial recognition unit; D 16 (i) represents the hexadecimal form of the identification information of the i-th recorded person stored in the storage and recording module; () 16 Indicates converting the value in the brackets to hexadecimal form; len[] indicates the number of bits of the hexadecimal data in the brackets; << indicates left shift; i = 1, 2, ..., n(t); n(t) indicates the total number of people recorded at the current moment; R{,} indicates judging the facial similarity of the data on both sides of the comma in the brackets. If it is judged to be the same face, the value 1 is output, and if it is judged to be not the same face, the value 0 is output;

[0065] The identification information generating unit is specifically used to generate identification information of a person according to the second formula;

[0066] Wherein, the second formula is:

[0067] D 16 [n(t)+1]=[n(t)+1] 16 < <len(G 16 )+G 16 (2)

[0068] In the second formula, D 16 [n(t)+1] indicates that when the facial feature information of the current person does not match any of the n(t) recorded persons, the identification information generating unit generates the identification information of the current person in hexadecimal form, and sends it to the storage and recording module so that it is stored and recorded as the identification information of the n(t)+1th recorded person; [] 16 Indicates converting the value in brackets to hexadecimal format.

[0069] The beneficial effects of the above technical solution are: using the first formula (1) to determine whether the person is a recorded person based on the face recognition situation of the current person, so as to perform corresponding control according to different face recognition states; using the second formula (2) to generate a recorded person number based on the face feature information of the person for storage and store its corresponding detection state, thereby forming a closed-loop calculation and control to ensure the reliability of the system.

[0070] As an optional embodiment, the authority determination unit includes:

[0071] a person determination subunit, configured to determine, when the judgment result of the first judgment unit is yes, a recorded person matching the facial feature information of the current person according to a third formula as a target recorded person;

[0072] A detection record acquisition subunit, used to acquire the historical detection records of the target record personnel stored in the storage record module;

[0073] An entry authority control value calculation subunit, used to calculate the entry authority control value of the current person at the current moment according to the fourth formula;

[0074] The judgment subunit is used to judge whether the access permission control value of the current person at the current moment is equal to 0. If so, it is determined to control closing the access permission of the current person to the target access control. Otherwise, it is determined to control releasing the access permission of the current person to the target access control.

[0075] Wherein, the third formula is:

[0076]

[0077] In the third formula, I indicates that the target recorded person determined at the current moment is the Ith recorded person; It means to substitute the value of i from 1 to n(t) into the brackets to obtain the value of i that makes the brackets true;

[0078] The fourth formula is:

[0079]

[0080] In the fourth formula, Q(t) represents the access control value of the current person at the current moment. If Q(t)=0, it is determined that the control closes the access right of the current person to the target access control. If Q(t)≠0, it is determined that the control releases the access right of the current person to the target access control. δ[] represents a judgment function. If the value in the bracket is 0, the function value is 1. If the value in the bracket is not 0, the function value is 0. J I (e) represents the human posture detection result recorded in the e-th historical detection record of the target recorded person. If the human posture of the target recorded person is detected to be normal in the e-th detection, then J I (e) = 0, otherwise (i.e., the target person’s body posture is detected to be abnormal for the eth time) J I (e)=1;e=1,2,…,B I ; B I It represents the total number of historical detection records of the target person acquired by the detection record acquisition subunit; J0 represents the body posture detection result of the current person.

[0081] The beneficial effect of the above technical solution is: the third formula (3) can be used to determine the recorded person who matches the facial feature information of the current person, so as to facilitate the subsequent corresponding control according to different facial recognition states, and then the fourth formula (4) is used to control the entry permission of the person according to whether the person is a recorded person and the corresponding detection state, thereby intelligently controlling the entry permission, reflecting the intelligence of the system.

[0082] As an optional embodiment, the human posture recognition security system based on edge computing further includes:

[0083] An alarm module is used to send an alarm message to a designated security monitoring terminal when the edge computing system determines to control and close the access rights of the current person to the target access control; wherein the alarm message includes the target access control identification, the walking image of the current person, and the access rights of the current person to the target access control determined by the edge computing system of the target access control;

[0084] The door opening control module is further used to control the target door to open when receiving the user door opening instruction returned by the designated security monitoring terminal according to the alarm information.

[0085] The beneficial effect of the above technical solution is that when there are illegal persons in front of the access control system, an alarm can be sounded through the alarm module, which is convenient for the access control system manager to timely understand the situation in front of the access control and take timely response measures, effectively improving the efficiency of access control system management.

[0086] It can be seen from the contents of the above embodiments that by adding an edge computing system 2 to the camera 1, the body posture of the human body can be detected, including the walking posture, the size of the steps, the amplitude of the arm swing, and whether the person has the right to enter; specifically: first, the face recognition of the human body is performed to determine whether the person is a recorded person. If the person is a recorded person, the historical detection status of the person is called up, and the entry permission of the person is controlled according to the historical detection status and the current detection status. If the person is not a recorded person, the entry permission of the person is controlled according to the current detection status. At the same time, a recorded person number is generated according to the facial information of the person for storage and the corresponding detection status is stored, which can effectively enhance security protection, expand the application scope of the access control system in some specific scenarios, and expand the diversity of security measures.

[0087] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 An apparatus for the method specified in one or more blocks.

[0088] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 The method specified in one or more boxes.

[0089] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps of the method are specified in one or more boxes.

[0090] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these changes and variations. The above are only specific embodiments 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 those skilled in the art within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A human posture recognition security system based on edge computing, It is characterized in that include: The camera is installed on the target access control and is used to collect the current walking image of the person; An edge computing system, arranged in the camera, is used to detect whether the body posture of the current person in the walking image of the current person is normal, and determine the access permission of the current person to the target access control according to a preset recognition algorithm and the body posture detection result of the current person; A door opening control module, connected to the edge computing system, for controlling the target access control to open when the edge computing system determines to release the access right of the current person to the target access control so as to allow the current person to enter; Wherein, the edge computing system includes: The posture detection unit is used to identify the human posture of the current person in the walking image of the current person, and detect whether the human posture of the current person is normal, and obtain the detection result; A face recognition unit is used to perform face recognition on the walking image of the current person to obtain the face feature information of the current person; A first judgment unit, configured to judge whether the facial feature information of the current person matches any recorded person according to the person detection record stored in the record storage module; an authority determination unit, for obtaining, when the judgment result of the first judgment unit is yes, a historical detection record of a recorded person that matches the facial feature information of the current person, and determining the access authority of the current person to the target access control system based on the currently obtained historical detection record and the human body posture detection result of the current person; an identification information generating unit, configured to generate identification information of the current person according to facial feature information of the current person when the judgment result of the first judgment unit is no, and send the identification information of the current person to the record storage module for storage as the identification information of the person being detected this time; The first judgment unit is specifically used to calculate the matching judgment value between the facial feature information of the current person at the current moment and the recorded person according to the first formula, and judge whether the matching judgment value between the facial feature information of the current person at the current moment and the recorded person is equal to 0, if so, judge that the facial feature information of the current person matches the recorded person, otherwise, judge that the facial feature information of the current person does not match the recorded person; Among them, the first formula is: In the first formula, E(t) represents the matching judgment value between the facial feature information of the current person at the current moment and the recorded person; t represents the current moment; G 16 The hexadecimal form of the facial feature information of the current person output by the facial recognition unit; D 16 (i) represents the hexadecimal form of the identification information of the i-th recorded person stored in the record storage module; () 16 Indicates converting the value in the brackets to hexadecimal form; len[] indicates the number of bits of the hexadecimal data in the brackets; << indicates left shift; i = 1, 2, ..., n(t); n(t) indicates the total number of people recorded at the current moment; R{,} indicates judging the facial similarity of the data on both sides of the comma in the brackets. If it is judged to be the same face, the value 1 is output, and if it is judged to be not the same face, the value 0 is output; The identification information generating unit is specifically used to generate identification information of a person according to the second formula; Wherein, the second formula is: D 16 [n(t)+1]=[n(t)+1] 16 <<len(G 16 )+G 16 In the second formula, D 16 [n(t)+1] indicates that when the facial feature information of the current person does not match any of the n(t) recorded persons, the identification information generating unit generates the identification information of the current person in hexadecimal form, and sends it to the record storage module so that it is stored and recorded as the identification information of the n(t)+1th recorded person; [] 16 Indicates converting the value in brackets to hexadecimal form; Wherein, the authority determination unit includes: a person determination subunit, configured to determine, when the judgment result of the first judgment unit is yes, a recorded person matching the facial feature information of the current person according to a third formula as a target recorded person; A test record acquisition subunit, used to acquire the historical test records of the target record personnel stored in the record storage module; An entry authority control value calculation subunit, used to calculate the entry authority control value of the current person at the current moment according to the fourth formula; The judgment subunit is used to judge whether the access permission control value of the current person at the current moment is equal to 0. If so, it is determined to control closing the access permission of the current person to the target access control. Otherwise, it is determined to control releasing the access permission of the current person to the target access control. Wherein, the third formula is: In the third formula, I indicates that the target recorded person determined at the current moment is the Ith recorded person; It means to substitute the value of i from 1 to n(t) into the brackets to obtain the value of i that makes the brackets true; The fourth formula is: In the fourth formula, Q(t) represents the access control value of the current person at the current moment; δ[] represents the 0 judgment function, if the value in the brackets is 0, the function value is 1, if the value in the brackets is not 0, the function value is 0; J I (e) represents the human posture detection result recorded in the e-th historical detection record of the target recorded person. If the human posture of the target recorded person is detected to be normal in the e-th detection, then J I (e) = 0, otherwise J I (e)=1;e=1,2,…,B I ; B I Indicates the total number of historical test records of the target record person acquired by the test record acquisition subunit; J 0 Indicates the body posture detection result of the current person.

2. The human posture recognition security system based on edge computing as claimed in claim 1, It is characterized in that The human body posture includes at least one of: the walking posture of a person, the size of the person's steps, and the amplitude of the person's arm swing; The edge computing system is specifically used to detect whether the human body posture of the current person in the walking image of the current person is normal according to a preset normal human body posture library.

3. The human posture recognition security system based on edge computing as claimed in claim 1, It is characterized in that The system further comprises: A record storage module is used to store the detection record of the detected person when the edge computing system determines to control and close the current person's access to the target door; the detection record includes the identification information of the detected person and the human body posture detection result of the detected person.

4. The human posture recognition security system based on edge computing as claimed in claim 1 or 3, It is characterized in that The system further comprises: An alarm module is used to send an alarm message to a designated security monitoring terminal when the edge computing system determines to control and close the access rights of the current person to the target access control; wherein the alarm message includes the target access control identification, the walking image of the current person, and the access rights of the current person to the target access control determined by the edge computing system of the target access control; The door opening control module is further used to control the target door to open when receiving the user door opening instruction returned by the designated security monitoring terminal according to the alarm information.

Images