A control method and system for electric sliding door of prison cell

Abstract

The present invention discloses a control method and system for an electric sliding door in a prison cell, and relates to the technical field of electric sliding door control, including analyzing the previous abnormal event records of each area to determine the risk of each area, thereby setting the opening risk threshold of each electric sliding door, and laying the foundation for automatic monitoring of electric sliding doors. The opening risk of the electric sliding door is determined according to the video information in the area and the video information in front of the door, and the relationship between the opening risk of the electric sliding door and the opening risk threshold is compared to obtain multiple situation types, and the opening, closing and opening degree of the electric sliding door are controlled for each situation type, thereby improving the adaptability and autonomy of the electric sliding door control, and when a dangerous situation is found according to the video stream, the control command of the electric sliding door is adjusted in time to ensure safe and stable operation in the prison.

Description

Technical Field

[0001] The invention relates to the technical field of electric sliding door control, and in particular to a control method and system for an electric sliding door of a prison cell. Background Art

[0002] In traditional prison management, the doors of cells are usually opened and closed manually, which has many disadvantages. First, manual door opening and closing is inefficient, time-consuming and labor-intensive, and increases the workload of prison staff. In the special environment of prisons, staff need to frequently enter and exit each cell for inspections, management and other work. Manual door opening and closing undoubtedly increases their physical exertion and time cost. Secondly, manual door opening and closing has hidden dangers in safety. Due to the uncertainty of human factors, such as negligence and fatigue, the door may not be closed or locked in time, posing risks to prison safety. In addition, manual door opening and closing is also easily affected by the external environment, such as strong winds and bad weather, which may cause the door to fail to close normally.

[0003] In contrast, the electric sliding door of the prison cell realizes the automatic opening and closing of the door through advanced control technology, which greatly improves the management efficiency and safety. In the existing technology, the electric sliding door can only identify the identity or open or close according to the instructions of the supervisor. However, in the event of sudden danger or critical incidents, the supervisor may not realize the danger in time or be in danger, and fail to control the electric sliding door in time, resulting in poor control adaptability and low control autonomy of the prison electric sliding door, which makes the impact of dangerous incidents continue to expand.

[0004] Therefore, how to improve the control adaptability and control autonomy of prison electric sliding doors is a technical problem that needs to be solved at present. Summary of the invention

[0005] The purpose of the present invention is to solve the problem of poor control adaptability and control autonomy of prison electric sliding doors in the prior art, and propose a control method for prison electric sliding doors, the method comprising:

[0006] Obtain geographic information within the prison, divide the prison into multiple different areas based on the geographic information within the prison, determine the electric sliding doors between multiple areas based on the connectivity between different areas, and thus build a matching relationship between areas and electric sliding doors;

[0007] Obtain the previous abnormal event records of each area, analyze the previous abnormal event records of each area to determine the risk of each area, set the opening risk threshold of the electric sliding door according to the matching relationship between the area and the electric sliding door and the risk of each area, obtain the surveillance video in each area and the surveillance video within a certain distance in front of each electric sliding door, and build the corresponding relationship between the surveillance video stream in the area and the surveillance video stream in front of the door;

[0008] Obtain the identity information of the supervisors and prisoners in the prison, determine the risk of opening the electric sliding door through the identity information of the supervisors and prisoners and the corresponding relationship between the surveillance video streams in the area and the surveillance video streams in front of the door, and compare the relationship between the opening risk of the electric sliding door and the opening risk threshold to divide multiple situation types;

[0009] The opening, closing and opening degree of the electric sliding door are controlled by the situation type and the number of people in the area, and the effective time of the prisoner's authority on the electric sliding door is set according to the daily activity flow of the prisoners, so as to complete the control of the electric sliding door.

[0010] In some embodiments of the present application, the electric sliding doors between multiple areas are determined based on the connectivity relationship between different areas, thereby establishing a matching relationship between the areas and the electric sliding doors, including:

[0011] Two adjacent and interconnected regions are regarded as regions with a connectivity relationship, and regions where the door type at the connection between the regions with a connectivity relationship is an electric sliding door are screened, and region-electric sliding door-region is regarded as a matching relationship between the region and the electric sliding door.

[0012] In some embodiments of the present application, the risk level of each area is determined by analyzing the previous abnormal event records of each area, including:

[0013] Collect records of past abnormal events in all areas and summarize all types of abnormal events in the prison;

[0014] If there is a jailbreak event among all abnormal event types in the prison, the jailbreak event will be used as the baseline event;

[0015] If there is no jailbreak event among all abnormal event types in the prison, one or more abnormal event types are selected from other abnormal event types as benchmark events;

[0016] Collect the records of previous benchmark events, calculate the average value of each evaluation dimension of the benchmark events, and assign different evaluation weights to each evaluation dimension through the average value of each evaluation dimension of the benchmark events, so as to calculate the severity of non-benchmark events and the severity of benchmark events, where non-benchmark events are abnormal event types other than benchmark events among abnormal events;

[0017] Determine the risk level of each region by using the attributes of the benchmark events and non-benchmark events in each region, including severity and frequency of occurrence;

[0018] ;

[0019] in, is the risk level of the ith region, is the risk conversion coefficient of the ith region, is the number of types of benchmark events in the ith region, is the first The combined weight of the benchmark events, is the first The frequency of occurrence of benchmark events, is the first The severity of the benchmark event, is an exponential function, is the number of types of non-benchmark events in the ith region, is the first The combined weight of non-benchmark events, is the first The frequency of non-benchmark events, is the first The severity of the non-baseline event, is the constant of the ith region.

[0020] In some embodiments of the present application, the risk threshold for opening the electric sliding door is set according to the matching relationship between the area and the electric sliding door and the risk of each area, including:

[0021] The importance of each area in the prison is evaluated according to its function, the electric sliding door corresponding to the two areas is determined according to the matching relationship between the area and the electric sliding door, and the opening risk threshold of the electric sliding door corresponding to the two areas is determined based on the importance and risk of the two areas.

[0022] In some embodiments of the present application, a corresponding relationship between the surveillance video stream in the area and the surveillance video stream in front of the door is constructed, including:

[0023] Determine the frame rates of the surveillance video stream in the area and the surveillance video stream in front of the door, and confirm the edges and center points of the surveillance video in the area and the surveillance video in front of the door, which are respectively recorded as the area edge, the area center point, the door edge and the door center point, and the door center point. According to the distance between the area edge and the door edge, the distance between the area edge and the door center point, the distance between the area center point and the door edge, and the distance between the area center point and the door center point, the actual distance range between the area and the door is mapped. The time difference between the surveillance video stream in the area and the surveillance video stream in front of the door is determined by the frame rates of the surveillance video stream in the area and the surveillance video stream in front of the door, and the actual distance range between the area and the door, and it is used as the correspondence between the two video streams.

[0024] In some embodiments of the present application, the risk of opening the electric sliding door is determined by the identity information of the supervisor and the prisoner and the corresponding relationship between the surveillance video stream in the area and the surveillance video stream in front of the door, including:

[0025] Through the identity information of supervisors and prisoners, face recognition is performed in the surveillance video stream and the surveillance video stream in front of the door, so as to determine the identity information of the people in the two video streams, and the video stream is intercepted according to the time difference between the surveillance video stream in the area and the surveillance video stream in front of the door. The difference between the continuous frames of the people in the video stream is compared through motion detection, so as to identify the dynamic behavior of the people;

[0026] Calculate the matching degree between the dynamic behavior of personnel and abnormal events. When the matching degree meets the requirements of abnormal events, it is determined that an abnormal event exists. The risk of opening the electric sliding door is determined based on the abnormal event.

[0027] When the matching degree does not meet the requirements of abnormal events, the transition probability between personnel dynamic behavior and abnormal events is calculated, and abnormal events with a transition probability greater than the transition probability threshold are screened out as candidate abnormal events. The risk of opening the electric sliding door is determined by the transition probability of the candidate abnormal events.

[0028] In some embodiments of the present application, the risk of opening the electric sliding door is determined by the transition probability of the abnormal event to be selected, including:

[0029] According to the surveillance video stream in the area and the surveillance video stream in front of the door, the transition probability of the dynamic behavior of the personnel and the selected abnormal event is divided into a first transition probability and a second transition probability, the first transition probability corresponds to the surveillance video stream in the area, and the second transition probability corresponds to the surveillance video stream in front of the door;

[0030] Determine the risk of opening the electric sliding door according to the first transfer probability and the second transfer probability;

[0031] ;

[0032] in, The risk of opening the electric sliding door. is the risk conversion factor, is the number of abnormal events corresponding to the first transition probability, For the The first transition probability of an abnormal event, is the transition probability threshold, is the weight function of the difference between the first transition probability and the transition probability threshold, For the The severity of an abnormal event, is the number of abnormal events corresponding to the second transition probability, For the The second transition probability of an abnormal event, is the weight function of the difference between the second transition probability and the transition probability threshold, For the The severity of an abnormal event.

[0033] In some embodiments of the present application, the relationship between the opening risk of the electric sliding door and the opening risk threshold is compared to divide multiple situation types, including:

[0034] The opening risk threshold includes a first threshold and a second threshold, and the first threshold is greater than the second threshold. The situation types include a high-risk situation, a low-risk situation, and a non-risk situation.

[0035] If the risk of opening the electric sliding door is greater than the first threshold, it is determined to be a high-risk situation;

[0036] If the risk of opening the electric sliding door is not greater than the first threshold and greater than the second threshold, it is determined to be a low-risk situation;

[0037] If the opening risk of the electric sliding door is not greater than the second threshold, it is determined to be a non-dangerous situation.

[0038] Correspondingly, the present application also provides a control system for an electric sliding door of a cell, comprising:

[0039] The first module is used to obtain geographic information within the prison, divide the prison into multiple different areas based on the geographic information within the prison, determine the electric sliding doors between multiple areas based on the connectivity between different areas, and thus build a matching relationship between areas and electric sliding doors;

[0040] The second module is used to obtain the previous abnormal event records of each area, analyze the previous abnormal event records of each area to determine the risk of each area, set the opening risk threshold of the electric sliding door according to the matching relationship between the area and the electric sliding door and the risk of each area, obtain the surveillance video in each area and the surveillance video within a certain distance in front of each electric sliding door, and construct the corresponding relationship between the surveillance video stream in the area and the surveillance video stream in front of the door;

[0041] The third module is used to obtain the identity information of the supervisors and prisoners in the prison, determine the risk of opening the electric sliding door through the identity information of the supervisors and prisoners and the corresponding relationship between the surveillance video stream in the area and the surveillance video stream in front of the door, and compare the relationship between the opening risk of the electric sliding door and the opening risk threshold to divide multiple situation types;

[0042] The fourth module is used to control the opening, closing and opening degree of the electric sliding door according to the situation type and the number of people in the area, and to set the effective time of the prisoner's authority on the electric sliding door according to the daily activity flow of the prisoners, so as to complete the control of the electric sliding door.

[0043] By applying the above technical solutions, the previous abnormal event records of each area are analyzed to determine the risk level of each area, and the opening risk threshold of each electric sliding door is set accordingly, laying the foundation for the automatic monitoring of electric sliding doors. The opening risk level of the electric sliding door is determined based on the video information in the area and the video information in front of the door, and the relationship between the opening risk level of the electric sliding door and the opening risk threshold is compared to obtain multiple scenario types. The opening, closing and opening degree of the electric sliding door are controlled for each scenario type, which improves the adaptability and autonomy of the electric sliding door control. When a dangerous situation is found according to the video stream, the control instructions of the electric sliding door are adjusted in time to ensure safe and stable operation in the prison. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A schematic diagram of a control method for an electric sliding door of a prison cell proposed by the present invention;

[0045] Figure 2 The present invention is a schematic structural diagram of a control system for an electric sliding door of a prison cell. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0047] Reference Figure 1A control method for an electric sliding door of a cell comprises the following steps:

[0048] Step S101, obtaining geographic information within the prison, dividing the prison into multiple different areas based on the geographic information within the prison, determining the electric sliding doors between the multiple areas based on the connectivity between the different areas, thereby building a matching relationship between the areas and the electric sliding doors.

[0049] In this embodiment, the electric sliding door is mainly composed of a door body, a drive system, a control system and a safety protection system. The door body is usually made of high-strength aluminum alloy or stainless steel, and has excellent corrosion resistance, wind pressure resistance and stability. The drive system converts electrical energy into mechanical energy through components such as motors, reducers, and transmission belts, and drives the door body to move in translation. The control system is responsible for receiving user instructions, controlling the operation of the door body, and realizing various safety protection functions. When a door opening signal is received (such as a person approaching, a remote control operation, or authorization by the access control system, etc.), the control system drives the motor to rotate forward, and the transmission belt drives the door body to move to one side to open the door. Similarly, when a door closing signal is received, the control system drives the motor to reverse, and the door body moves to the other side to close the door. During the entire process, the control system will monitor the state of the door body in real time to ensure its smooth and safe operation.

[0050] In some embodiments of the present application, the electric sliding doors between multiple areas are determined based on the connectivity relationship between different areas, thereby establishing a matching relationship between the areas and the electric sliding doors, including:

[0051] Two adjacent and interconnected regions are regarded as regions with a connectivity relationship, and regions where the door type at the connection between the regions with a connectivity relationship is an electric sliding door are screened, and region-electric sliding door-region is regarded as a matching relationship between the region and the electric sliding door.

[0052] It is understandable that if an electric sliding door connects more than two areas, it also belongs to the areas with a connected relationship.

[0053] Step S102, obtain the previous abnormal event records of each area, analyze the previous abnormal event records of each area to determine the risk of each area, set the opening risk threshold of the electric sliding door according to the matching relationship between the area and the electric sliding door and the risk of each area, obtain the surveillance video in each area and the surveillance video within a certain distance in front of each electric sliding door, and construct the correspondence between the surveillance video stream in the area and the surveillance video stream in front of the door.

[0054] In this embodiment, abnormal events include non-compliant or emergencies such as prisoner violence, prisoner escape, prisoner-supervisor conflict, and fire, etc., so as to assign a risk level to each area and determine the risk level threshold for opening the electric sliding door. The situation in the monitoring area and the situation within a certain distance in front of the door (mainly to prevent events such as coercion identification, such as prisoners holding supervisors hostage to open special electric sliding doors that only supervisors have the authority to open).

[0055] In some embodiments of the present application, the risk level of each area is determined by analyzing the previous abnormal event records of each area, including:

[0056] Collect records of past abnormal events in all areas and summarize all types of abnormal events in the prison;

[0057] If there is a jailbreak event among all abnormal event types in the prison, the jailbreak event will be used as the baseline event;

[0058] If there is no jailbreak event among all abnormal event types in the prison, one or more abnormal event types are selected from other abnormal event types as benchmark events;

[0059] Collect the records of previous benchmark events, calculate the average value of each evaluation dimension of the benchmark events, and assign different evaluation weights to each evaluation dimension through the average value of each evaluation dimension of the benchmark events, so as to calculate the severity of non-benchmark events and the severity of benchmark events, where non-benchmark events are abnormal event types other than benchmark events among abnormal events;

[0060] Determine the risk level of each region by using the attributes of the benchmark events and non-benchmark events in each region, including severity and frequency of occurrence;

[0061] .

[0062] in, is the risk level of the ith region, is the risk conversion coefficient of the ith region, is the number of types of benchmark events in the ith region, is the first The combined weight of the benchmark events, is the first The frequency of occurrence of benchmark events, is the first The severity of the benchmark event, is an exponential function, is the number of types of non-benchmark events in the ith region, is the first The combined weight of non-benchmark events, is the first The frequency of non-benchmark events, is the first The severity of the non-baseline event, is the constant of the ith region.

[0063] In this embodiment, generally speaking, the purpose of a prison is to prevent prisoners from escaping. Therefore, if there has been a jailbreak in history (the result of escaping is not necessary, as long as there is an act of escaping), the jailbreak will be used as the benchmark event. If there has been no jailbreak in history, it means that the prison's security management and education are probably good, and other events with greater impact will be used as benchmark events. The benchmark event is the event with the greatest impact in the prison by default. Other events are quantified through the benchmark event, and the evaluation dimensions include casualties, property losses, and the degree of threat to prison security and order.

[0064] In this embodiment, it means that the influence of the reference event is corrected by the non-reference event, and the number of reference events is at least 1.

[0065] In some embodiments of the present application, the risk threshold for opening the electric sliding door is set according to the matching relationship between the area and the electric sliding door and the risk of each area, including:

[0066] The importance of each area in the prison is evaluated according to its function, the electric sliding door corresponding to the two areas is determined according to the matching relationship between the area and the electric sliding door, and the opening risk threshold of the electric sliding door corresponding to the two areas is determined based on the importance and risk of the two areas.

[0067] In this embodiment, the importance and risk of the two areas are combined to determine the opening risk threshold of the electric sliding door corresponding to the two areas. The safety level of an electric sliding door can be determined by weighted summation or averaging, and different safety levels correspond to two different opening risk thresholds.

[0068] In some embodiments of the present application, a corresponding relationship between the surveillance video stream in the area and the surveillance video stream in front of the door is constructed, including:

[0069] Determine the frame rates of the surveillance video stream in the area and the surveillance video stream in front of the door, and confirm the edges and center points of the surveillance video in the area and the surveillance video in front of the door, which are respectively recorded as the area edge, the area center point, the door edge and the door center point, and the door center point. According to the distance between the area edge and the door edge, the distance between the area edge and the door center point, the distance between the area center point and the door edge, and the distance between the area center point and the door center point, the actual distance range between the area and the door is mapped. The time difference between the surveillance video stream in the area and the surveillance video stream in front of the door is determined by the frame rates of the surveillance video stream in the area and the surveillance video stream in front of the door, and the actual distance range between the area and the door, and it is used as the correspondence between the two video streams.

[0070] In this embodiment, the situation in the area and the identification situation in front of the door are monitored by the video in the area and the video within a certain distance in front of the door. Because there is a certain distance between the two, there is a certain time difference when the two video streams are proofread. It takes a certain distance and time for the object in the area to walk to the door. For the monitoring video in the area and the monitoring video in front of the door, we need to confirm their edges and center points respectively. This can be achieved through geometric analysis of the video screen, such as using image processing technology to identify the edges and center points of the screen. For the monitoring video stream in the area and the monitoring video stream in front of the door, we need to determine their frame rates respectively. The frame rate refers to the number of frames displayed per second, which determines the time resolution of events in the video stream. With the frame rate and the actual distance range, we can calculate the time difference between the monitoring video stream in the area and the monitoring video stream in front of the door. The calculation of the time difference can be based on a simple proportional relationship: the time difference can be approximated as the ratio of the actual distance range to the moving distance per second corresponding to the frame rate.

[0071] Step S103, obtaining the identity information of the supervisors and prisoners in the prison, determining the risk of opening the electric sliding door through the identity information of the supervisors and prisoners and the correspondence between the surveillance video stream in the area and the surveillance video stream in front of the door, and comparing the relationship between the opening risk of the electric sliding door and the opening risk threshold to divide into multiple situation types.

[0072] In this embodiment, a deep learning algorithm (such as YOLO, SSD, etc.) is used to detect objects of interest (prisoners and supervisors, etc.) in the video frame. The events that occur in the area and in front of the door are judged to evaluate the risk of opening the electric sliding door. The larger the value, the greater the risk of opening.

[0073] In some embodiments of the present application, the risk of opening the electric sliding door is determined by the identity information of the supervisor and the prisoner and the corresponding relationship between the surveillance video stream in the area and the surveillance video stream in front of the door, including:

[0074] Through the identity information of supervisors and prisoners, face recognition is performed in the surveillance video stream and the surveillance video stream in front of the door, so as to determine the identity information of the people in the two video streams, and the video stream is intercepted according to the time difference between the surveillance video stream in the area and the surveillance video stream in front of the door. The difference between the continuous frames of the people in the video stream is compared through motion detection, so as to identify the dynamic behavior of the people;

[0075] Calculate the matching degree between the dynamic behavior of personnel and abnormal events. When the matching degree meets the requirements of abnormal events, it is determined that an abnormal event exists. The risk of opening the electric sliding door is determined based on the abnormal event.

[0076] When the matching degree does not meet the requirements of abnormal events, the transition probability between personnel dynamic behavior and abnormal events is calculated, and abnormal events with a transition probability greater than the transition probability threshold are screened out as candidate abnormal events. The risk of opening the electric sliding door is determined by the transition probability of the candidate abnormal events.

[0077] In this embodiment, the matching degree between the dynamic behavior of personnel and the abnormal event is calculated. When the matching degree meets the requirements of the abnormal event, it is determined that an abnormal event exists. The risk of opening the electric sliding door is determined according to the abnormal event. It is determined whether an abnormal event has occurred according to the matching degree between the dynamic behavior of personnel and the abnormal event. If an abnormal event has occurred, the risk of opening the electric sliding door corresponding to the abnormal event is analyzed. The severity of the abnormal event is considered here. For example, if an event of prisoners attacking supervisors has occurred, it can be understood as a jailbreak. As long as this event exists, the risk of opening the electric sliding door will be relatively large. If no abnormal event has occurred, the transition probability between the dynamic behavior of personnel and the abnormal event is calculated. Here, the state transition probability can be calculated through the Markov chain model, that is, the probability of transferring from the dynamic behavior of personnel to the abnormal event state, and the abnormal events with a transition probability greater than the transition probability threshold are screened out. If there is no abnormal event with a transfer probability greater than the transfer probability threshold, it is determined to be a non-dangerous situation.

[0078] In some embodiments of the present application, the risk of opening the electric sliding door is determined by the transition probability of the abnormal event to be selected, including:

[0079] According to the surveillance video stream in the area and the surveillance video stream in front of the door, the transition probability of the dynamic behavior of the personnel and the selected abnormal event is divided into a first transition probability and a second transition probability, the first transition probability corresponds to the surveillance video stream in the area, and the second transition probability corresponds to the surveillance video stream in front of the door;

[0080] Determine the risk of opening the electric sliding door according to the first transfer probability and the second transfer probability;

[0081] .

[0082] in, The risk of opening the electric sliding door. is the risk conversion factor, is the number of abnormal events corresponding to the first transition probability, For the The first transition probability of an abnormal event, is the transition probability threshold, is the weight function of the difference between the first transition probability and the transition probability threshold, For the The severity of an abnormal event, is the number of abnormal events corresponding to the second transition probability, For the The second transition probability of an abnormal event, is the weight function of the difference between the second transition probability and the transition probability threshold, For the The severity of an abnormal event.

[0083] In this embodiment, the first transition probability and the second transition probability are the transition probability of the abnormal event in the area with higher transition probability after screening and the transition probability of the abnormal event in front of the door, respectively. is the first transition probability of the th abnormal event in the area, is the second transition probability of the th abnormal event in front of the door, and is the same.

[0084] In some embodiments of the present application, the relationship between the opening risk of the electric sliding door and the opening risk threshold is compared to divide multiple situation types, including:

[0085] The opening risk threshold includes a first threshold and a second threshold, and the first threshold is greater than the second threshold. The situation types include a high-risk situation, a low-risk situation, and a non-risk situation.

[0086] If the risk of opening the electric sliding door is greater than the first threshold, it is determined to be a high-risk situation;

[0087] If the risk of opening the electric sliding door is not greater than the first threshold and greater than the second threshold, it is determined to be a low-risk situation;

[0088] If the opening risk of the electric sliding door is not greater than the second threshold, it is determined to be a non-dangerous situation.

[0089] In this embodiment, the high-risk situation is a relatively dangerous situation, such as a prisoner escaping from prison, etc., and the low-risk situation is a relatively low-risk situation, such as a fight between prisoners. The non-dangerous situation is a non-dangerous situation, and the electric sliding door is opened or closed normally.

[0090] Step S104, controlling the opening, closing and opening degree of the electric sliding door by the situation type and the number of people in the area, and setting the effective time of the prisoner's authority on the electric sliding door according to the daily activity flow of the prisoners, thereby completing the control of the electric sliding door.

[0091] In this embodiment, for high-risk situations, the opening and closing of the electric sliding door are controlled according to the type of abnormal event. For example, if a prisoner escapes from prison in the area, the corresponding electric sliding door is closed to prevent the prisoner from escaping until the supervisor resolves the incident and opens it. For low-risk situations, for example, if there is a fight between prisoners in the area, the door can be controlled to be opened or closed. Depending on the handling strategy of the supervisor for similar incidents, the door can be closed to reduce the number of prisoners coming in to watch the excitement and reduce the impact. The opening degree of the door can be controlled to appropriately reduce the flow of people, while ensuring the normal entry of prisoners, which is convenient for supervisors to adjust. For non-dangerous situations, the opening degree of the door is dynamically adjusted according to the number of people. In addition to fully opening, the opening degree here can also be set to 1 / 4 or 1 / 2 of the door. For some doors, prisoners can be given the right to open, but the time is set. For example, the door of the cell can only be opened within a specific time.

[0092] By applying the above technical solutions, the previous abnormal event records of each area are analyzed to determine the risk level of each area, and the opening risk threshold of each electric sliding door is set accordingly, laying the foundation for the automatic monitoring of electric sliding doors. The opening risk level of the electric sliding door is determined based on the video information in the area and the video information in front of the door, and the relationship between the opening risk level of the electric sliding door and the opening risk threshold is compared to obtain multiple scenario types. The opening, closing and opening degree of the electric sliding door are controlled for each scenario type, which improves the adaptability and autonomy of the electric sliding door control. When a dangerous situation is found according to the video stream, the control instructions of the electric sliding door are adjusted in time to ensure safe and stable operation in the prison.

[0093] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present invention can be implemented by hardware, or by software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present invention.

[0094] Correspondingly, the present application also provides a control system for an electric sliding door of a cell, such as Figure 2 As shown, including,

[0095] The first module is used to obtain geographic information within the prison, divide the prison into multiple different areas based on the geographic information within the prison, determine the electric sliding doors between multiple areas based on the connectivity between different areas, and thus build a matching relationship between areas and electric sliding doors;

[0096] The second module is used to obtain the previous abnormal event records of each area, analyze the previous abnormal event records of each area to determine the risk of each area, set the opening risk threshold of the electric sliding door according to the matching relationship between the area and the electric sliding door and the risk of each area, obtain the surveillance video in each area and the surveillance video within a certain distance in front of each electric sliding door, and construct the corresponding relationship between the surveillance video stream in the area and the surveillance video stream in front of the door;

[0097] The third module is used to obtain the identity information of the supervisors and prisoners in the prison, determine the risk of opening the electric sliding door through the identity information of the supervisors and prisoners and the corresponding relationship between the surveillance video stream in the area and the surveillance video stream in front of the door, and compare the relationship between the opening risk of the electric sliding door and the opening risk threshold to divide multiple situation types;

[0098] The fourth module is used to control the opening, closing and opening degree of the electric sliding door according to the situation type and the number of people in the area, and to set the effective time of the prisoner's authority on the electric sliding door according to the daily activity flow of the prisoners, so as to complete the control of the electric sliding door.

[0099] Those skilled in the art will appreciate that the accompanying drawings are merely schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily required for implementing the present invention.

[0100] Those skilled in the art will appreciate that the modules in the system in the implementation scenario can be distributed in the system of the implementation scenario according to the implementation scenario description, or can be changed accordingly and located in one or more systems different from the implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple submodules.

[0101] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A control method for an electric sliding door of a prison cell, characterized in that: include, Obtain geographic information within the prison, divide the prison into multiple different areas based on the geographic information within the prison, determine the electric sliding doors between multiple areas based on the connectivity between different areas, and thus build a matching relationship between areas and electric sliding doors; Obtain the previous abnormal event records of each area, analyze the previous abnormal event records of each area to determine the risk of each area, set the opening risk threshold of the electric sliding door according to the matching relationship between the area and the electric sliding door and the risk of each area, obtain the surveillance video in each area and the surveillance video within a certain distance in front of each electric sliding door, and build the corresponding relationship between the surveillance video stream in the area and the surveillance video stream in front of the door; Obtain the identity information of the supervisors and prisoners in the prison, determine the risk of opening the electric sliding door through the identity information of the supervisors and prisoners and the corresponding relationship between the surveillance video streams in the area and the surveillance video streams in front of the door, and compare the relationship between the opening risk of the electric sliding door and the opening risk threshold to divide multiple situation types; The opening, closing and opening degree of the electric sliding door are controlled by the situation type and the number of people in the area, and the effective time of the prisoner's authority on the electric sliding door is set according to the daily activity flow of the prisoners, so as to complete the control of the electric sliding door; Among them, the electric sliding doors between multiple areas are determined according to the connectivity relationship between different areas, so as to construct a matching relationship between the area and the electric sliding door, including: Two adjacent and interconnected regions are regarded as regions with a connectivity relationship, and regions where the door type at the connection point between the regions with a connectivity relationship is an electric sliding door are selected, and region-electric sliding door-region is regarded as a matching relationship between the region and the electric sliding door; Analyze the past abnormal event records of each area to determine the risk level of each area, including, Collect records of past abnormal events in all areas and summarize all types of abnormal events in the prison; If there is a jailbreak event among all abnormal event types in the prison, the jailbreak event will be used as the baseline event; If there is no jailbreak event among all abnormal event types in the prison, one or more abnormal event types are selected from other abnormal event types as benchmark events; Collect the records of previous benchmark events, calculate the average value of each evaluation dimension of the benchmark events, and assign different evaluation weights to each evaluation dimension through the average value of each evaluation dimension of the benchmark events, so as to calculate the severity of non-benchmark events and the severity of benchmark events, where non-benchmark events are abnormal event types other than benchmark events among abnormal events; Determine the risk level of each region by using the attributes of the benchmark events and non-benchmark events in each region, including severity and frequency of occurrence; ； in, is the risk level of the ith region, is the risk conversion coefficient of the ith region, is the number of types of benchmark events in the ith region, is the first The combined weight of the benchmark events, is the first The frequency of occurrence of benchmark events, is the first The severity of the benchmark event, is an exponential function, is the number of types of non-benchmark events in the ith region, is the first The combined weight of non-benchmark events, is the first The frequency of non-baseline events, is the first The severity of the non-baseline event, is the constant of the ith region; The risk threshold for opening the electric sliding door is set based on the matching relationship between the area and the electric sliding door and the risk level of each area, including: According to the functions of each area in the prison, the importance of each area is evaluated, and the corresponding electric sliding door between two areas is determined according to the matching relationship between the area and the electric sliding door. The opening risk threshold of the electric sliding door corresponding to the two areas is determined by combining the importance and risk of the two areas. And build the corresponding relationship between the surveillance video stream in the area and the surveillance video stream in front of the door, including, Determine the frame rates of the surveillance video stream in the area and the surveillance video stream in front of the door, and confirm the edges and center points of the surveillance video in the area and the surveillance video in front of the door, which are recorded as the area edge, the area center point, the door edge and the door center point respectively. Map the actual distance range between the area and the door according to the distance between the area edge and the door edge, the distance between the area edge and the door center point, the distance between the area center point and the door edge, and the distance between the area center point and the door center point. Determine the time difference between the surveillance video stream in the area and the surveillance video stream in front of the door according to their frame rates and the actual distance range between the area and the door, and use it as the corresponding relationship between the two video streams. The risk of opening the electric sliding door is determined by the identity information of the supervisor and the prisoner, and the corresponding relationship between the surveillance video stream in the area and the surveillance video stream in front of the door, including: Through the identity information of supervisors and prisoners, face recognition is performed in the surveillance video stream and the surveillance video stream in front of the door, so as to determine the identity information of the people in the two video streams, and the video stream is intercepted according to the time difference between the surveillance video stream in the area and the surveillance video stream in front of the door. The difference between the continuous frames of the people in the video stream is compared through motion detection, so as to identify the dynamic behavior of the people; Calculate the matching degree between the dynamic behavior of personnel and abnormal events. When the matching degree meets the requirements of abnormal events, it is determined that an abnormal event exists. The risk of opening the electric sliding door is determined based on the abnormal event. When the matching degree does not meet the requirements of abnormal events, the transition probability between the personnel dynamic behavior and the abnormal event is calculated, and the abnormal events with a transition probability greater than the transition probability threshold are screened out as candidate abnormal events. The risk of opening the electric sliding door is determined by the transition probability of the candidate abnormal events. By comparing the relationship between the opening risk of the electric sliding door and the opening risk threshold, multiple scenario types are divided, including: The opening risk threshold includes a first threshold and a second threshold, and the first threshold is greater than the second threshold. The situation types include a high-risk situation, a low-risk situation, and a non-risk situation; If the risk of opening the electric sliding door is greater than the first threshold, it is determined to be a high-risk situation; If the risk of opening the electric sliding door is not greater than the first threshold and greater than the second threshold, it is determined to be a low-risk situation; If the opening risk of the electric sliding door is not greater than the second threshold, it is determined to be a non-dangerous situation.

2. The control method of the electric sliding door of the prison cell according to claim 1 is characterized in that: The risk of opening the electric sliding door is determined by the transfer probability of the selected abnormal events, including: According to the surveillance video stream in the area and the surveillance video stream in front of the door, the transition probability of the dynamic behavior of the personnel and the selected abnormal event is divided into a first transition probability and a second transition probability, the first transition probability corresponds to the surveillance video stream in the area, and the second transition probability corresponds to the surveillance video stream in front of the door; Determine the risk of opening the electric sliding door according to the first transfer probability and the second transfer probability; ； in, The risk of opening the electric sliding door. is the risk conversion factor, is the number of abnormal events corresponding to the first transition probability, For the The first transition probability of an abnormal event, is the transition probability threshold, is the weight function of the difference between the first transition probability and the transition probability threshold, For the The severity of an abnormal event, is the number of abnormal events corresponding to the second transition probability, For the The second transition probability of an abnormal event, is the weight function of the difference between the second transition probability and the transition probability threshold, For the The severity of an abnormal event.

3. A control system for an electric sliding door of a prison cell, used to implement a control method for an electric sliding door of a prison cell as described in right 1, characterized in that: The system includes, The first module is used to obtain geographic information within the prison, divide the prison into multiple different areas based on the geographic information within the prison, determine the electric sliding doors between multiple areas based on the connectivity between different areas, and thus build a matching relationship between areas and electric sliding doors; The second module is used to obtain the previous abnormal event records of each area, analyze the previous abnormal event records of each area to determine the risk of each area, set the opening risk threshold of the electric sliding door according to the matching relationship between the area and the electric sliding door and the risk of each area, obtain the surveillance video in each area and the surveillance video within a certain distance in front of each electric sliding door, and construct the corresponding relationship between the surveillance video stream in the area and the surveillance video stream in front of the door; The third module is used to obtain the identity information of the supervisors and prisoners in the prison, determine the risk of opening the electric sliding door through the identity information of the supervisors and prisoners and the corresponding relationship between the surveillance video stream in the area and the surveillance video stream in front of the door, and compare the relationship between the opening risk of the electric sliding door and the opening risk threshold to divide multiple situation types; The fourth module is used to control the opening, closing and opening degree of the electric sliding door according to the situation type and the number of people in the area, and to set the effective time of the prisoner's authority on the electric sliding door according to the daily activity flow of the prisoners, so as to complete the control of the electric sliding door.

Images