Safety monitoring device based on artificial intelligence terminal control box

Abstract

The invention provides a safety monitoring device based on an artificial intelligence terminal control box. The safety monitoring device comprises an infrared sensing device, an access control camera device, an access control voice prompt device, an access control voice recording device, a control box, an intra-field camera device and an alarm. The method comprises the following steps: firstly, monitoring and recognizing the face of a person to enter a site, judging whether the person is a worker in the site or not, if so, directly releasing the person, if not, judging the person as a foreign person and not allowing the person to release, then judging whether the foreign person knows the related worker in the site or not, judging that the foreign person really knows and allows the person to release, and otherwise, leaving the person outside the entrance guard of the site. Secondly, whether the foreign personnel in the field are safe personnel or dangerous personnel is judged, when the action track of the foreign personnel conforms to the reasonable action track corresponding to the identity of the foreign personnel, even if the foreign personnel are not the internal personnel in the field, the foreign personnel are still allowed to act in the field area, the convenience of management of the internal personnel in the field can be improved, and identification is flexible.

Description

Technical Field

[0001] The present invention relates to the technical field of security monitoring, and in particular to a security monitoring device based on an artificial intelligence terminal control box. Background Art

[0002] The construction site refers to the approved construction site where industrial and civil projects such as housing construction, civil engineering, equipment installation, pipeline laying and other construction activities are carried out, and the place where people carry out safe production, civilized work and construction. The project manager is fully responsible for the on-site management of the construction process. According to the scale of the project, the technical complexity and the specific conditions of the construction site, the construction site management responsibilities should be established and organized and implemented.

[0003] The existing construction site area may have security issues for the entry of outsiders. Although the existing intelligent monitoring system can automatically identify the entry of outsiders, the identification mode is single. For people entering the site, single-point monitoring is often adopted. The movement trajectory of people entering the site cannot be monitored as a whole, and it is difficult to accurately identify high-risk people among visitors, which poses a major safety hazard. Summary of the invention

[0004] In order to solve the above technical problems, a security monitoring device based on an artificial intelligence terminal control box is provided. This technical solution solves the problem that the existing intelligent monitoring system proposed in the above background technology can automatically identify the entry of external personnel, but the identification mode is single. For personnel entering the site, a single-point monitoring method is often adopted, and the movement trajectory of personnel entering the site cannot be monitored as a whole. It is difficult to accurately identify high-risk personnel among visitors, and there are major safety hazards.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A security monitoring device based on an artificial intelligence terminal control box comprises an infrared sensing device, a door control camera device, a door control voice prompting device, a door control voice recording device, a control box, an on-site camera device and an alarm, wherein the infrared sensing device, the door control camera device, the door control voice prompting device, the door control voice recording device, the on-site camera device and the alarm are all electrically connected to the control box, and the infrared sensing device is also electrically connected to the door control camera device, the infrared sensing device, the door control camera device, the door control voice prompting device and the door control voice recording device are all arranged at the door control of the construction site, the on-site camera device is arranged in key areas, important facilities and main public activity places in the site, the control box is arranged at the gate guard of the construction site, the broadcasting device is arranged in several groups, which are respectively installed at the door control and key positions inside the construction site, the control box is installed with a control module, a door control cloud management module and an on-site cloud management module, and the front side of the control box is provided with a charging port and a data connector.

[0007] Preferably, the infrared sensing device monitors the surrounding environment and controls the switch of the access control camera device at the access control entrance through the control module.

[0008] Preferably, the access control cloud management module is internally installed with:

[0009] An access control photo analysis unit, which is used to obtain multiple groups of photos collected by the access control camera device and extract facial information from the photos;

[0010] A facial data storage unit for personnel on the field, the facial data storage unit for personnel on the field is used to store facial information of all personnel on the field;

[0011] The access control face determination unit is used to compare the face information in the photo with the face information of all staff members in the on-site face data storage unit to determine whether they are on-site staff members. If the faces match, they are allowed to pass directly.

[0012] Preferably, the access control cloud management module is also equipped with:

[0013] An on-site personnel telephone voice storage unit, the on-site personnel telephone voice storage unit is used to store the voice data in the telephone recordings of all the staff members in the factory;

[0014] Access control sound determination unit, the control module controls the switch of the access control voice prompt device, the access control voice prompt device sends out voice prompts to remind the outsiders to enter whether they know the relevant staff in the venue, and prompts the outsiders to call the relevant staff in the venue. If they do not know them, they are not allowed to pass; if they know them, after the call is connected, the control module controls the switch of the access control voice recording device to record the telephone conversation, and the access control sound determination unit makes a judgment based on the sound data in the telephone sound storage unit of the inside personnel. If the sound matches, they are directly released.

[0015] Preferably, the on-site cloud management module is internally installed with:

[0016] An on-site image monitoring unit, which obtains facial image information and movement trajectory from images captured by an on-site camera device;

[0017] The on-site image determination unit is electrically connected to the on-site image monitoring unit, and the on-site image determination unit determines whether the person is a safe person or a dangerous person based on the facial image information and movement trajectory in the image captured by the on-site camera device and using a comprehensive determination algorithm.

[0018] Preferably, the comprehensive determination algorithm specifically includes:

[0019] Based on the facial image data, the identity of the person entering the venue is determined, and based on the identity of the person entering the venue, their reasonable movement trajectory points are generated.

[0020] Based on the action trajectory collected by the video surveillance unit in the venue, combined with the reasonable action trajectory points, determine whether the action trajectory of the person entering the venue belongs to his reasonable action trajectory. If so, no response is made. If not, the unreasonable action trajectory points in his action trajectory are screened out to determine whether the unreasonable action trajectory points belong to the key areas in the venue. If so, a warning signal is directly output to the alarm to alarm. If not, determine whether the unreasonable action trajectory points belong to important facilities and major public activity places in the venue. If so, calculate the visitor's abnormal behavior index according to the abnormal behavior calculation logic to obtain the visitor's abnormal behavior index value. If not, no response is made.

[0021] Determine whether the visitor's abnormal behavior index value is greater than the abnormal behavior preset value. If so, output a warning signal to the alarm to alarm. If not, do not respond.

[0022] Preferably, the specific formula for calculating the visitor's abnormal behavior index is:

[0023] A=N1×γ1+N2×γ2

[0024] Among them, A is the visitor's abnormal behavior index value, N1 is the number of unreasonable action trajectory points belonging to important facilities in the venue, N2 is the number of unreasonable action trajectory points belonging to major public activity places in the venue, γ1 is the calculation weight of important facilities in the venue, and γ2 is the calculation weight of major public activity places in the venue.

[0025] Compared with the prior art, the present invention provides a security monitoring device based on an artificial intelligence terminal control box, which has the following beneficial effects:

[0026] 1. The present invention first monitors and recognizes the face of the person to be admitted, and determines whether he is a staff member in the venue. If so, he is directly released. If not, he is determined to be an outsider and is not allowed to be released. Then the access control voice prompt device sends a voice prompt to remind the outsider to enter whether he knows the relevant staff in the venue, and prompts the outsider to call the relevant staff in the venue. If he knows, the outsider needs to call the relevant staff he knows in the venue. The access control voice recording device records the conversation, and the voice storage unit of the telephone of the personnel in the venue stores the voice information of a telephone recording of all the staff in the factory area. The access control voice determination unit compares the recording of this conversation with the voice in the voice storage unit of the telephone of the personnel in the venue, and determines that the outsider does know the relevant staff in the venue and is allowed to be released. Otherwise, he is left outside the venue access control.

[0027] 2. The present invention secondly determines whether the outsider inside the venue is a safe person or a dangerous person. When the movement trajectory of the outsider conforms to the reasonable movement trajectory corresponding to his identity, even if he is not an insider in the venue, he is still allowed to appear in the venue area, which can improve the convenience of internal personnel management in the venue. When a single high-risk event or multiple medium- and high-risk events occur, a warning signal is directly output to the alarm to alarm, which is handled by the security management personnel, thereby effectively ensuring the safety of the venue and flexible identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a schematic diagram of the principle structure of the present invention;

[0029] Figure 2 It is a structural schematic diagram of the control box in the present invention;

[0030] Figure 3 It is a schematic diagram of the internal structure of the control box in the present invention;

[0031] Figure 4 It is a schematic diagram of the internal structure of the access control cloud management module in the present invention;

[0032] Figure 5 It is a schematic diagram of the internal structure of the in-field cloud management module in the present invention.

[0033] In the figure: 1. Infrared sensor; 2. Access control camera; 3. Access control voice prompt device; 4. Access control voice recording device; 5. Control box; 501. Charging port; 502. Data connector; 503. Control module; 504. Access control cloud management module; 505. On-site cloud management module; 506. Access control photo analysis unit; 507. On-site personnel face data storage unit; 508. Access control face determination unit; 509. On-site personnel telephone voice storage unit; 510. Access control voice determination unit; 511. On-site image analysis unit; 512. On-site image determination unit; 6. On-site camera device; 7. Alarm. DETAILED DESCRIPTION

[0034] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific implementation methods of the present invention are now described with reference to the accompanying drawings, but the protection scope of the present invention is not limited to the following.

[0035] Please refer to Figure 1-5 As shown, a security monitoring device based on an artificial intelligence terminal control box includes an infrared sensing device 1, a door control camera device 2, a door control voice prompt device 3, a door control voice recording device 4, a control box 5, an on-site camera device 6 and an alarm 7. The infrared sensing device 1, the door control camera device 2, the door control voice prompt device 3, the door control voice recording device 4, the on-site camera device 6 and the alarm 7 are all electrically connected to the control box 5, and the infrared sensing device 1 is also electrically connected to the door control camera device 2. The voice prompt device 3 and the access control voice recording device 4 are both installed at the access control of the construction site. The on-site camera device 6 is installed in key areas, important facilities and major public activity places on the site. The control box 5 is installed at the gate guard of the construction site. The broadcasting device 7 is provided with several groups, which are respectively installed at the access control and key positions inside the construction site. The control box 5 is installed with a control module 503, an access control cloud management module 504 and an on-site cloud management module 505. The front side of the control box 5 is provided with a charging port 501 and a data connector 502.

[0036] The infrared sensor device 1 monitors the surrounding environment and controls the switch of the access control camera device 2 at the access control entrance through the control module 503 .

[0037] The access control cloud management module 504 is internally installed with:

[0038] The access control photo analysis unit 506 is used to obtain multiple groups of photos collected by the access control camera device 2 and extract facial information from the photos;

[0039] The on-site personnel face data storage unit 507 is used to store the face information of all the staff members on the site;

[0040] The access control face determination unit 508 is used to compare the face information in the photo with the face information of all staff members in the on-site face data storage unit 507 to determine whether they are on-site staff members. If the faces match, they are allowed to pass directly.

[0041] The access control cloud management module 504 also has the following components installed inside:

[0042] On-site personnel telephone voice storage unit 509, the on-site personnel telephone voice storage unit 509 is used to store the voice data in the telephone recordings of all the staff members in the factory;

[0043] The access control sound determination unit 510 and the control module 503 control the switch of the access control voice prompt device 3. The access control voice prompt device 3 issues a voice prompt to remind the outsiders to enter whether they know the relevant staff in the venue, and prompt the outsiders to call the relevant staff in the venue. If they do not know them, they are not allowed to pass. If they know them, after the call is connected, the control module 503 controls the switch of the access control voice recording device 4 to record the telephone conversation, and the access control sound determination unit 510 makes a judgment based on the sound data in the telephone sound storage unit 509 of the inside personnel. If the sound matches, they are directly released.

[0044] In this scheme, the personnel to be admitted are first monitored and face recognized to determine whether they are staff members in the venue. If so, they are directly released. If not, they are determined to be outsiders and are not allowed to be released. Then the access control voice prompt device 3 issues a voice prompt to remind the outsiders to be admitted whether they know the relevant staff members in the venue, and prompts the outsiders to call the relevant staff members in the venue. If they know them, the outsiders need to call the relevant staff members they know in the venue. The access control voice recording device 4 records the conversation, and the voice storage unit 509 for the telephone of the personnel in the venue stores the voice information of a telephone recording of all the staff members in the factory area. The access control voice determination unit 510 compares the recording of this conversation with the voice in the voice storage unit 509 for the telephone of the personnel in the venue to determine that the outsiders do know the relevant staff members in the venue and are allowed to be released. Otherwise, they are kept outside the venue access control.

[0045] The internal installation of the on-site cloud management module 505 includes:

[0046] An in-field image monitoring unit 511, which obtains facial image information and movement trajectory from the image captured by the in-field camera device 6;

[0047] The in-field image determination unit 512 is electrically connected to the in-field image monitoring unit 511, and the in-field image determination unit 512 determines whether the person is a safe person or a dangerous person based on the facial image information and movement trajectory in the image captured by the in-field camera device 6, and uses a comprehensive determination algorithm.

[0048] The comprehensive judgment algorithm specifically includes:

[0049] Based on the facial image data, determine the identity of the person entering the venue, and generate reasonable movement trajectory points based on the identity of the person entering the venue;

[0050] Based on the action trajectory collected by the on-site image monitoring unit 511, combined with the reasonable action trajectory points, it is determined whether the action trajectory of the person entering the venue belongs to his reasonable action trajectory. If so, no response is made. If not, the unreasonable action trajectory points in his action trajectory are screened out to determine whether the unreasonable action trajectory points belong to the key areas in the venue. If so, a warning signal is directly output to the alarm 7 for alarm. If not, it is determined whether the unreasonable action trajectory points belong to the important facilities and main public activity places in the venue. If so, the visitor abnormal behavior index is calculated according to the abnormal behavior calculation logic to obtain the visitor abnormal behavior index value. If not, no response is made.

[0051] Determine whether the visitor's abnormal behavior index value is greater than the abnormal behavior preset value. If so, output a warning signal to the alarm 7 for alarm. If not, do not respond.

[0052] The specific formula for calculating the visitor's abnormal behavior index is:

[0053] A=N1×γ1+N2×γ2

[0054] Among them, A is the visitor's abnormal behavior index value, N1 is the number of unreasonable action trajectory points belonging to important facilities in the venue, N2 is the number of unreasonable action trajectory points belonging to major public activity places in the venue, γ1 is the calculation weight of important facilities in the venue, and γ2 is the calculation weight of major public activity places in the venue.

[0055] In this solution, when the movement trajectory of an outsider is consistent with the reasonable movement trajectory corresponding to his identity, even if he is not an internal staff in the venue, he is still allowed to appear in the venue area, which can improve the convenience of internal personnel management in the venue. When a single high-risk event or multiple medium- and high-risk events occur, a warning signal is directly output to the alarm 7 for alarm, which will be handled by the security management personnel, thereby effectively ensuring the safety of the venue and flexible identification.

[0056] The working principle and use process of the present invention are as follows: first, the personnel to be admitted are monitored and face recognition is performed to determine whether they are on-site staff. If so, they are directly released. If not, they are determined to be outsiders and are not allowed to be released. Then, the access control voice prompt device 3 issues a voice prompt to remind the outsiders to be admitted whether they know the relevant staff on the site, and prompts the outsiders to call the relevant staff on the site. If they know, the outsiders need to call the relevant staff they know on the site. The access control voice recording device 4 records the conversation, and the on-site personnel telephone sound storage unit 509 stores the sound information in a telephone recording of all the staff on the factory area. The access control sound determination unit 510 performs voice recognition on the recorded conversation and the sound in the on-site personnel telephone sound storage unit 509. By comparison, it is determined that the outsider does know the relevant staff on site and is allowed to pass; otherwise, he is kept outside the site access control; then it is determined whether the outsider on site is a safe person or a dangerous person. When the movement trajectory of the outsider conforms to the reasonable movement trajectory corresponding to his identity, even if he is not an internal staff member on site, he is still allowed to appear in the internal area of ​​the site, which can improve the convenience of internal personnel management on site. When a single high-risk event or multiple medium- and high-risk events occur, a warning signal is directly output to the alarm 7 for alarm, which is handled by the security management personnel, thereby effectively ensuring the safety of the site and flexible in identification. The present invention is not only suitable for the monitoring of construction sites, but also can be used in other places in life, such as factories, schools, hospitals, etc., which improves the applicability of the present invention.

[0057] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention still fall within the scope covered by the present invention.

Claims

1. A security monitoring device based on an artificial intelligence terminal control box, characterized in that: The invention comprises an infrared sensor device (1), a door control camera device (2), a door control voice prompt device (3), a door control voice recording device (4), a control box (5), an on-site camera device (6) and an alarm (7), wherein the infrared sensor device (1), the door control camera device (2), the door control voice prompt device (3), the door control voice recording device (4), the on-site camera device (6) and the alarm (7) are all electrically connected to the control box (5), and the infrared sensor device (1) is also electrically connected to the door control camera device (2). ) and the access control voice recording device (4) are both arranged at the access control of the construction site, the on-site camera device (6) is arranged at key areas, important facilities and major public activity places on the site, the control box (5) is arranged at the gate guard of the construction site, the broadcast device (7) is arranged in several groups, which are respectively installed at the access control and key positions inside the construction site, the control box (5) is installed with a control module (503), an access control cloud management module (504) and an on-site cloud management module (505), and the front side of the control box (5) is provided with a charging port (501) and a data connector (502).

2. A security monitoring device based on an artificial intelligence terminal control box according to claim 1, characterized in that: The infrared sensing device (1) monitors the surrounding environment and controls the switch of the access control camera device (2) at the access control entrance through the control module (503).

3. The security monitoring device based on the artificial intelligence terminal control box according to claim 1 is characterized in that: The access control cloud management module (504) is internally installed with: A door access photo analysis unit (506), the door access photo analysis unit (506) is used to obtain multiple groups of photos collected by the door access camera device (2) and extract facial information from the photos; An on-site personnel face data storage unit (507), the on-site personnel face data storage unit (507) is used to store the face information of all the staff members on the site; The access control face determination unit (508) is used to compare the face information in the photo with the face information of all staff members in the on-site face data storage unit (507) to determine whether they are on-site staff members. If the faces match, they are directly allowed to pass.

4. The security monitoring device based on the artificial intelligence terminal control box according to claim 1 is characterized in that: The access control cloud management module (504) is also equipped with: An on-site personnel telephone voice storage unit (509), the on-site personnel telephone voice storage unit (509) is used to store the voice data in the telephone recordings of all the staff members in the factory; The door control sound determination unit (510) controls the switch of the door control voice prompt device (3). The door control voice prompt device (3) issues a voice prompt to remind the outsider who is about to enter whether he knows the relevant staff in the venue, and prompts the outsider to call the relevant staff in the venue. If he does not know the staff, he is not allowed to pass. If he knows the staff, after the call is connected, the control module (503) controls the switch of the door control voice recording device (4) to record the telephone conversation. The door control sound determination unit (510) makes a judgment based on the sound data in the telephone sound storage unit (509) of the inside personnel. If the sound matches, the outsider is directly allowed to pass.

5. The security monitoring device based on the artificial intelligence terminal control box according to claim 1 is characterized in that: The on-site cloud management module (505) is internally installed with: An on-site image monitoring unit (511), the on-site image monitoring unit (511) acquiring facial image information and movement trajectory of a person in an image captured by an on-site camera device (6); The in-field image determination unit (512) is electrically connected to the in-field image monitoring unit (511), and the in-field image determination unit (512) determines whether the person is a safe person or a dangerous person based on the facial image information and movement trajectory in the image captured by the in-field camera device (6) and using a comprehensive determination algorithm.

6. A security monitoring device based on an artificial intelligence terminal control box according to claim 5, characterized in that: The comprehensive determination algorithm specifically includes: Based on the facial image data, determine the identity of the person entering the venue, and generate reasonable movement trajectory points based on the identity of the person entering the venue; Based on the action trajectory collected by the on-site image monitoring unit (511), combined with the reasonable action trajectory points, it is determined whether the action trajectory of the person entering the venue belongs to his reasonable action trajectory. If so, no response is made. If not, the unreasonable action trajectory points in his action trajectory are screened out to determine whether the unreasonable action trajectory points belong to the key areas of the venue. If so, a warning signal is directly output to the alarm (7) for alarm. If not, it is determined whether the unreasonable action trajectory points belong to the important facilities and main public activity places in the venue. If so, the visitor abnormal behavior index is calculated according to the abnormal behavior calculation logic to obtain the visitor abnormal behavior index value. If not, no response is made. It is determined whether the visitor's abnormal behavior index value is greater than the abnormal behavior preset value. If so, a warning signal is output to the alarm (7) for alarm. If not, no response is made.

7. The security monitoring device based on the artificial intelligence terminal control box according to claim 7 is characterized in that: The specific formula for calculating the visitor's abnormal behavior index is: A=N1×γ1+N2×γ2 Among them, A is the visitor's abnormal behavior index value, N1 is the number of unreasonable action trajectory points belonging to important facilities in the venue, N2 is the number of unreasonable action trajectory points belonging to major public activity places in the venue, γ1 is the calculation weight of important facilities in the venue, and γ2 is the calculation weight of major public activity places in the venue.

Images