Intelligent alarm triggering device and fireproof door

Through the multi-module combination of intelligent alarm triggering devices, the fire door status and environmental safety level are identified in real time and early warning instructions are generated, which solves the problem of insufficient predictability of traditional fire door systems and improves the prevention and response capabilities of fire hazards.

CN118570960BActive Publication Date: 2024-11-15WANGDOU JIU AN DOOR CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202410623200.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-20
Publication Date
2024-11-15
Estimated Expiration
2044-05-20

AI Technical Summary

Technical Problem

Traditional fire doors and alarm systems are insufficient in preventive protection of fire hazards and have limited protection performance.

Method used

An intelligent alarm triggering device was designed, through the stairwell positioning module, door status recognition module, floor status information acquisition module, environmental safety identification module and safety level matching module, combined with camera monitoring video and sensor data, the fire door status and real-time identification and Environmental safety level, generate early warning instructions.

Benefits of technology

It has achieved forward-looking triggering, improved the fire safety protection level of buildings, and enhanced the prediction and response capabilities of fire hazards.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118570960B_ABST
    Figure CN118570960B_ABST
Patent Text Reader

Abstract

The present invention discloses an intelligent alarm triggering device and a fire door, and relates to the technical field of signal alarm. The method comprises: a stairwell positioning module obtains a building structure design drawing, extracts a plurality of stairwells of a target building, including a plurality of floor fire door sets; a door state recognition module retrieves a plurality of monitoring video sets to perform fire door state recognition, and obtains a plurality of first safety factor sets; a floor state information acquisition module collects floor state information and floor evacuation partition information sets; an environmental safety recognition module performs environmental safety recognition, and obtains a plurality of second safety factor sets; a safety level matching module performs fire safety level matching based on a plurality of first safety factor sets and a plurality of second safety factor sets, and obtains a plurality of fire safety levels; and an early warning module, wherein the early warning module is used to perform early warnings on a plurality of stairwells based on a plurality of fire safety levels, and generate early warning instructions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of signal alarms, and in particular to an intelligent alarm triggering device and a fireproof door. Background Art

[0002] A fire door is a door that serves as a fire-proof isolation inside a building. A fire alarm device is usually installed outside the door. Fire doors and fire alarm devices are important components of fire safety in modern buildings.

[0003] Traditional fire doors and alarm systems mainly rely on manual operation or simple automation devices. Although these devices can provide fire protection to a certain extent, their performance is often unsatisfactory when faced with preventive protection situations of fire hazards. There are technical problems such as insufficient predictability and limited protective performance. Summary of the invention

[0004] The present invention provides an intelligent alarm triggering device and a fireproof door to solve the technical problems of insufficient predictability and limited protective performance in the prior art, and to achieve the technical effects of forward-looking triggering and improving the safety protection level.

[0005] In a first aspect, the present invention provides an intelligent alarm triggering device, wherein the device comprises:

[0006] A stairwell positioning module, the stairwell positioning module is used to obtain a building structure design drawing of a target building, and extract multiple stairwells of the target building based on the building structure design drawing, wherein the multiple stairwells have multiple positioning building location identifiers, and the multiple stairwells include multiple floor fire door sets;

[0007] A door status recognition module, the door status recognition module is used to retrieve multiple monitoring video sets of multiple camera sets of the multiple floor fire door sets within a preset monitoring period, perform fire door status recognition on the multiple monitoring video sets, and obtain multiple first safety factor sets;

[0008] A floor status information collection module, the floor status information collection module is used to collect floor status information and floor evacuation partition information set of the target building within a preset monitoring period;

[0009] An environmental safety identification module, the environmental safety identification module is used to perform environmental safety identification on a plurality of floor fire door sets of the target building according to the floor status information and the floor evacuation partition information set, and obtain a plurality of second safety factor sets;

[0010] A safety level matching module, the safety level matching module is used to perform fire safety level matching on the multiple stairwells based on the multiple first safety factor sets and the multiple second safety factor sets to obtain multiple fire safety levels;

[0011] An early warning module is used to issue early warnings to the sensor monitoring data of the multiple stairwells based on the multiple fire safety levels and generate early warning instructions.

[0012] In a second aspect, the present invention further provides an intelligent alarm triggering method, wherein the method comprises:

[0013] Acquire a building structure design drawing of a target building, and extract multiple stairwells of the target building based on the building structure design drawing, wherein the multiple stairwells have multiple positioning building location identifiers, and the multiple stairwells include multiple floor fire door sets;

[0014] Retrieving multiple monitoring video sets of multiple camera sets of the fire door sets on the multiple floors within a preset monitoring period, identifying the states of the fire doors on the multiple monitoring video sets, and obtaining multiple first safety factor sets;

[0015] Collecting floor status information and floor evacuation partition information set of the target building within a preset monitoring period;

[0016] Performing environmental safety identification on multiple floor fire door sets of the target building according to the floor status information and the floor evacuation partition information set to obtain multiple second safety factor sets;

[0017] Performing fire safety level matching on the plurality of stairwells based on the plurality of first safety factor sets and the plurality of second safety factor sets to obtain a plurality of fire safety levels;

[0018] Based on the multiple fire safety levels, early warning is performed on the sensor monitoring data of the multiple stairwells to generate early warning instructions.

[0019] The present invention discloses an intelligent alarm triggering device and a fire door, comprising: a stairwell positioning module extracts the stairwell of a target building through a building structure design drawing, and identifies the location mark and the floor fire door. The door status recognition module uses the camera monitoring video to identify the state of the fire door and generates a first safety factor. The floor status information acquisition module collects the floor status and evacuation partition information. The environmental safety recognition module performs environmental safety recognition on the floor fire door according to the floor information and generates a second safety factor. The safety level matching module matches the fire safety level of the stairwell according to the safety factor. The early warning module warns the monitoring data of the stairwell according to the fire safety level and generates an early warning instruction. The intelligent alarm triggering device and fire door disclosed by the present invention solve the technical problems of insufficient predictability and limited protection performance, and achieve the technical effect of forward-looking triggering and improving the level of safety protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a structural schematic diagram of an intelligent alarm triggering device of the present invention;

[0021] Figure 2 The figure is a flow chart of an intelligent alarm triggering method of the present invention.

[0022] Explanation of the reference numerals: stairwell positioning module 11 , door status recognition module 12 , floor status information acquisition module 13 , environmental safety recognition module 14 , safety level matching module 15 , early warning module 16 . DETAILED DESCRIPTION

[0023] The technical solution provided in the embodiments of the present invention is to solve the technical problems of high latency, lack of real-time performance, incomplete monitoring, and insufficient predictability in the prior art. The overall idea adopted is as follows:

[0024] The stairwell positioning module first obtains the architectural structure design drawing of the target building, and then extracts multiple stairwells based on the design drawing. These stairwells have multiple positioning building location identifiers, including multiple floor fire door sets. Next, the door status recognition module retrieves multiple monitoring video sets of multiple camera sets of multiple floor fire door sets within a preset monitoring period, and performs fire door status recognition on these videos to obtain multiple first safety factor sets. The floor status information acquisition module is used to collect the floor status information and floor evacuation partition information sets of the target building within a preset monitoring period. The environmental safety recognition module performs environmental safety recognition on multiple floor fire door sets based on the floor status information and floor evacuation partition information sets to obtain multiple second safety factor sets. Finally, the early warning module issues early warnings on the sensor monitoring data of multiple stairwells based on multiple fire safety levels and generates early warning instructions.

[0025] The above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods of the specification to better understand the above technical solution. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments used only to explain the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, it should be noted that, for the convenience of description, only the parts related to the present invention are shown in the drawings, rather than all of them.

[0026] Embodiment 1

[0027] Figure 1 The present invention is a schematic diagram of the structure of an intelligent alarm triggering device; wherein, it includes:

[0028] A stairwell positioning module 11 is used to obtain a building structure design drawing of a target building, and extract multiple stairwells of the target building based on the building structure design drawing, wherein the multiple stairwells have multiple positioning building location identifiers, and the multiple stairwells include multiple floor fire door sets;

[0029] Specifically, first, obtain the architectural structure design drawing of the target building, which contains the internal structural design information of the target building. In other words, the architectural structure design drawing refers to the design drawing containing the overall structural layout, floor distribution, fire door location and other information of the target building.

[0030] Optionally, the location and layout of the stairwell are identified and extracted from the building structure design drawing, so as to obtain multiple stairwells of the target building, and the multiple stairwells have multiple positioning building location identifiers, and the multiple stairwells include multiple floor fire door sets. Among them, the positioning building location identifier refers to a specific identifier or mark used to identify the location of the stairwell. Floor fire doors refer to fire doors located on different floors. Multiple floor fire door sets refer to a structured set containing all floor fire doors formed by classifying multiple floor fire doors according to floors.

[0031] Through the above steps, the stairwells and floor fire doors of the target building are accurately identified and classified, providing the necessary information for the intelligent alarm triggering system. This extraction method based on the building structure design drawing can improve the accuracy and response speed of the alarm triggering system, helping to ensure the safety of people in the building.

[0032] The door state recognition module 12 is used to retrieve multiple monitoring video sets of multiple camera sets of the fire door sets on the multiple floors within a preset monitoring period, perform fire door state recognition on the multiple monitoring video sets, and obtain multiple first safety factor sets;

[0033] The preset monitoring period refers to a pre-set time period for monitoring and data collection. The camera set refers to a camera installed in a target building for monitoring the status of fire doors. The monitoring video set refers to a set of monitoring video data collected by multiple cameras in the above camera set within the preset monitoring period.

[0034] Optionally, based on image processing technology and machine vision algorithms, the fire door status is identified in the surveillance video collection, including whether the fire door is closed, open, damaged, and other abnormal conditions.

[0035] For example, first, obtain the camera surveillance video within a preset monitoring period through the network or storage device. Decode and process the video, and extract the image frames for subsequent analysis. Then, use computer vision technology to analyze and process the extracted image frames, identify the fire door in the image, and determine whether it is open or closed. Specifically, it involves technical means such as object detection, image segmentation and feature extraction.

[0036] Optionally, the corresponding safety factor is calculated according to the state of the fire door. For example, if the fire door is in a closed state, the safety factor is higher; if it is in an open state, the safety factor is lower.

[0037] In some embodiments, the door status identification module performs the following steps:

[0038] Constructing a state recognition network layer, wherein the state recognition network layer includes a first subject recognition branch, a first obstacle recognition branch, and a safety factor recognition branch;

[0039] The state recognition network layer is used to perform image analysis on the multiple monitoring video sets respectively to obtain the multiple first safety factor sets.

[0040] Optionally, the door state recognition module 12 performs image analysis of multiple surveillance video sets through a constructed state recognition network layer, thereby obtaining multiple first safety factor sets. Among them, the first subject recognition branch is used to identify the subject in the surveillance video, that is, the fire door. The first obstacle recognition branch is used to identify obstacles that may exist in front of the fire door, such as piled up debris or objects that hinder the opening or closing of the fire door. The safety factor recognition branch is used to calculate the safety factor of the fire door based on the state of the subject and the obstacle.

[0041] Optionally, the first subject recognition branch uses target detection or instance segmentation techniques to achieve subject recognition and positioning. The output of this branch can provide location information of the fire door for subsequent state judgment and safety factor calculation.

[0042] Optionally, the first obstacle recognition branch detects and recognizes obstacles using techniques such as object detection or image segmentation. The output of this branch can help determine the unobstructed state of the fire door and whether there are safety hazards.

[0043] Optionally, the safety factor identification branch provides an assessment of the safety status of the fire door and a basis for triggering an alarm based on the open and closed state of the fire door, the conditions of the surrounding environment and possible obstacles.

[0044] Through the combination and collaborative work of the above branches, the state recognition network layer can conduct a comprehensive analysis of the surveillance video and derive a corresponding set of safety factors. The above set of safety factors helps the system to promptly detect and deal with safety issues inside the building.

[0045] In some embodiments, the door state identification module further comprises:

[0046] Collecting multiple sample images of fire doors on multiple sample floors at different opening and closing angles, wherein the multiple sample images have multiple angle identifiers;

[0047] Using a convolutional network layer to perform supervised learning on the mapping relationship between the multiple sample images and the multiple angle identifiers, and updating the network parameters according to the output results until the output reaches convergence, thereby obtaining the first subject recognition branch that has been trained;

[0048] Using the first subject recognition branch to dynamically recognize video frames in the plurality of surveillance video sets, respectively, and output a plurality of angle change sequence sets;

[0049] The multiple angle change sequence sets are respectively used as multiple first identification data sets, and are associated and mapped with multiple second identification data sets obtained after the first obstacle identification branch performs obstacle identification on the multiple surveillance video sets, to obtain multiple safety identification data sets;

[0050] The plurality of safety identification data sets are analyzed in the safety factor identification branch to obtain the plurality of first safety factor sets.

[0051] Optionally, based on a convolutional neural network, a convolutional neural network layer is constructed for supervised learning, and the ability to recognize the state of the fire door is acquired through the mapping relationship between multiple sample images and multiple angle identifiers.

[0052] Specifically, first, multiple sample images of fire doors on multiple sample floors at different opening and closing angles are collected. These sample images have multiple angle identifiers, which are used to train the first subject recognition branch based on the convolutional network layer. Then, the convolutional network layer is used to supervise the mapping relationship between the sample images and the angle identifiers, and the network parameters are updated according to the output results until the output converges. In this way, the system can obtain a trained first subject recognition branch, and after training, the first subject recognition branch can accurately identify fire doors at different opening and closing angles.

[0053] Furthermore, the trained first subject recognition branch is used to dynamically recognize the video frames in the surveillance video, and multiple angle change sequence sets are output. The multiple angle change sequence sets are used as multiple first recognition data sets, and are associated and mapped with multiple second recognition data sets after the first obstacle recognition branch performs obstacle recognition on multiple surveillance video sets, thereby obtaining multiple safety recognition data sets. Among them, the second recognition data set refers to a data set containing obstacle recognition results. The angle change sequence is associated with the obstacle recognition result, so as to comprehensively consider the state of the fire door and the surrounding environment, and obtain a comprehensive safety recognition data set.

[0054] Optionally, the safety identification data set is analyzed in the safety factor identification branch to calculate multiple first safety factor sets. Exemplarily, the safety factor identification branch analyzes the safety status by evaluating indicators of the safety status in the safety identification data set, such as the opening angle of the fire door, the distance between the obstacle and the fire door, the angle of the obstacle interfering with the opening and closing space of the fire door, etc.

[0055] In some embodiments, the door state identification module further comprises:

[0056] Collecting a plurality of sample angle change sequences, a plurality of sample second identification data and a plurality of sample first safety factors as construction data;

[0057] The constructed data is used as a plurality of analysis points in a two-dimensional analysis space, wherein a one-dimensional coordinate axis of the two-dimensional analysis space is a sample angle variation sequence, a two-dimensional coordinate axis is second identification data, and each analysis point has a sample first safety factor identifier;

[0058] The two-dimensional analysis space is embedded in the safety factor identification branch, and the multiple safety identification data sets are analyzed to obtain the multiple first safety factor sets.

[0059] Optionally, a safety factor identification branch is constructed based on multiple sample angle change sequences, multiple sample second identification data and multiple sample first safety factors.

[0060] Specifically, firstly, multiple sample angle change sequences, multiple sample second identification data and multiple sample first safety factors are collected as construction data. Then, the construction data is mapped to multiple analysis points in the two-dimensional analysis space. The one-dimensional coordinate axis of the two-dimensional analysis space is the sample angle change sequence, which represents the change in the opening and closing angle of the fire door; the two-dimensional coordinate axis is the second identification data, which represents the state change of the environment around the fire door; each analysis point has a sample first safety factor identifier, which represents the safety factor of the point. In this way, the system can intuitively represent and analyze the construction data in the two-dimensional analysis space. Next, the two-dimensional analysis space is embedded in the safety factor identification branch, and multiple safety identification data sets are analyzed to obtain multiple first safety factor sets.

[0061] Through the above configuration steps, a comprehensive analysis and evaluation of the fire door status can be achieved, providing key information for building safety management, improving the accuracy and intelligence level of safety management, and helping to ensure the safety of people in the building.

[0062] In some embodiments, the door state identification module further comprises:

[0063] Extracting the first security identification data from the plurality of security identification data sets, inputting the first security identification data into the two-dimensional analysis space, and obtaining a first point to be identified;

[0064] Calculate the Euclidean distances from a plurality of analysis points in the two-dimensional analysis space to the first point to be identified, and arrange the calculation results in ascending order, and use the first m analysis points as the target point set;

[0065] Performing weighted calculation on the first safety factor identifiers of the m samples of the target point set to generate a first safety factor of the first point to be identified;

[0066] The multiple safety identification data sets are analyzed using the two-dimensional analysis space to obtain the multiple first safety factor sets.

[0067] Exemplarily, first, extract the first security identification data from multiple security identification data sets, and use the extracted first security identification data as input, input into the two-dimensional analysis space, and obtain the first point to be identified. Then, calculate the Euclidean distance from multiple analysis points in the two-dimensional analysis space to the first point to be identified. And arrange the calculation results in order from small to large, and obtain the first m analysis points closest to the first point to be identified as the target point set. Among them, m is determined based on the safety control level of the target building, and the higher the safety control level, the larger m. Then, based on the Euclidean distance from multiple analysis points to the first point to be identified, a weighted calculation is performed on the first safety factor identification of the first m samples of the target point set, wherein the weight is proportional to the inverse of the distance, that is, the closer the Euclidean distance from the analysis point to the first point to be identified, the greater the weight of the analysis point. The result of the weighted calculation is the first safety factor of the first point to be identified, thereby obtaining multiple first safety factor sets:

[0068] Through the above steps, the target point set can be weighted calculated to obtain the first safety factor of the first point to be identified, and multiple safety identification data sets can be analyzed using a two-dimensional analysis space to obtain multiple first safety factor sets to evaluate the safety status of fire doors and provide a scientific basis for building safety management.

[0069] A floor status information collection module 13 is used to collect floor status information and floor evacuation partition information set of the target building within a preset monitoring period;

[0070] Among them, floor status information refers to the status information of each floor of the target building during the monitoring period, such as whether there is someone on the floor, whether an abnormal event occurs on the floor, etc. Floor evacuation zone information refers to the zone information of each floor of the target building for evacuation in an emergency, including evacuation paths, evacuation instructions, etc. The above information and information sets can be collected through various sensors (such as crowd flow sensors, temperature sensors, etc.) and monitoring equipment (such as cameras, access control systems, etc.).

[0071] In some embodiments, the execution steps of the floor status information collection module include:

[0072] Counting the pedestrian flow data of multiple floors of the target building within a preset monitoring period to obtain the pedestrian flow data of multiple floors;

[0073] respectively comparing the plurality of floor traffic data with the sum of the plurality of floor traffic data to obtain a plurality of floor density coefficients, and using the plurality of floor density coefficients as floor status information;

[0074] Extracting the evacuation guidance signs of the target building, and determining multiple evacuation planning area sets of multiple floor fire door sets;

[0075] Based on the multiple evacuation planning area sets, the floor evacuation zone information set is generated, and each floor evacuation zone information corresponds to a floor of the target building.

[0076] Optionally, a video surveillance system or sensor or other equipment is used to monitor and record the entry and exit of people in the building in real time, and obtain the flow data of people on each floor of the target building within a preset monitoring period. By counting the flow of people on each floor, the distribution of people and the density of people can be understood.

[0077] Optionally, after obtaining the flow data of people on each floor, calculate the floor density coefficient. The floor density coefficient is calculated by dividing the flow of people on each floor by the sum of the flow of people on all floors. This coefficient reflects the relative value of the density of people on each floor and is one of the important indicators for evaluating the density of people on a floor.

[0078] Among them, the evacuation planning area refers to the floor area allocated to a certain evacuation passage or evacuation exit (fire door) in an emergency, and the evacuation planning area is determined based on the evacuation guidance signs. And the evacuation guidance signs are marked on the building floor plan. Specifically, the method of extracting the evacuation planning area includes manual annotation, image processing, and pattern recognition technologies. The extracted evacuation planning area can help determine the proportion of evacuation tasks undertaken by each fire door plan on the current floor.

[0079] An environmental safety identification module 14 is used to perform environmental safety identification on a plurality of floor fire door sets of the target building according to the floor status information and the floor evacuation partition information set, and obtain a plurality of second safety factor sets;

[0080] In some embodiments, the execution step of the environment safety identification module further includes:

[0081] Allocating the multiple floor density coefficient sets according to the floor evacuation partition information set to generate multiple fire door evacuation pressure coefficient sets corresponding to the multiple floor fire door sets;

[0082] The fire door evacuation pressure coefficients in the plurality of fire door evacuation pressure coefficient sets are respectively divided by the sum of the coefficients of the plurality of fire door evacuation pressure coefficient sets to obtain a plurality of second safety factor sets.

[0083] Optionally, based on the correspondence between floors, multiple floor density coefficient sets are allocated according to the floor evacuation partition information set, and multiple fire door evacuation pressure coefficient sets corresponding to the multiple floor fire door sets are obtained. The fire door evacuation pressure coefficient reflects the level of human flow allocated to the fire door. The larger the fire door evacuation pressure coefficient, the larger the evacuation human flow that will be allocated to the corresponding fire door.

[0084] Optionally, the result of dividing the evacuation pressure coefficient of each fire door by the sum of the evacuation pressure coefficients of all fire doors is calculated to obtain the evacuation pressure coefficient ratio of each fire door as multiple second safety factor sets, and the second safety factor sets reflect the proportion of evacuation flow that may be allocated to the fire door in an emergency. The method based on the calculation of the evacuation pressure coefficient can improve the accuracy and response speed of the security management system, and help to ensure the safety of people in the building.

[0085] A safety level matching module 15, configured to perform fire safety level matching on the plurality of stairwells based on the plurality of first safety factor sets and the plurality of second safety factor sets to obtain a plurality of fire safety levels;

[0086] Optionally, based on the safety level preset in the safety control plan of the target building, a first safety factor set and a second safety factor range corresponding to the safety level are configured to obtain a safety response matrix. The matrix is ​​used to guide safety measures and emergency responses taken at different safety levels.

[0087] Exemplarily, the following safety levels and corresponding safety responses are defined: High safety level corresponds to stairwells with a high first safety factor and a low second safety factor, and conventional safety management measures can be maintained. Medium safety level corresponds to stairwells with a moderate first safety factor and a moderate second safety factor, and it may be necessary to strengthen safety inspections, strengthen safety training and drills, etc. Low safety level corresponds to stairwells with a low first safety factor and a large second safety factor, and safety measures such as adding fire-fighting facilities, clearing obstacles, and strengthening inspections and monitoring may be required.

[0088] The security response matrix can be adjusted and customized according to the actual situation and security needs of the building to ensure that appropriate security measures can be taken in a timely and effective manner in an emergency to protect the safety of people inside the building.

[0089] Optionally, fire safety level matching is performed based on the above safety response matrix to obtain multiple fire safety levels, providing managers with real-time and accurate fire safety level information so as to quickly and effectively conduct emergency management and ensure the safety of the park.

[0090] The early warning module 16 is used to issue early warnings to the sensor monitoring data of the multiple stairwells based on the multiple fire safety levels and generate early warning instructions.

[0091] Optionally, the sensing monitoring data is acquired through a monitoring sensor. Preferably, the monitoring sensor includes a smoke sensor, an infrared sensor, a gas sensor, etc.

[0092] Optionally, the fire safety level corresponds to different warning thresholds, wherein a stairwell with a higher fire safety level corresponds to a lower warning threshold, that is, a stairwell with a higher fire safety level corresponds to a monitoring warning with a higher sensitivity.

[0093] In summary, the intelligent alarm triggering device provided by the present invention has the following technical effects:

[0094] The stairwell positioning module extracts multiple stairwells through the architectural design drawing. Each stairwell has multiple location markers and floor fire doors. The door status recognition module obtains the monitoring video, identifies the fire door status, and obtains the first safety factor. The floor status information acquisition module collects the floor status and evacuation information. The environmental safety recognition module identifies the environmental safety of the stairwell based on the information and obtains the second safety factor. The safety level matching module matches the fire safety level based on the two safety factors. The early warning module warns the sensor data based on the safety level and generates an early warning instruction. This achieves the technical effect of forward-looking triggering and improving the level of safety protection.

[0095] Embodiment 2

[0096] Figure 2 : is a flow chart of an intelligent alarm triggering method of the present invention. For example, Figure 1 The structural diagram of an intelligent alarm triggering device of the present invention can be shown as follows: Figure 2 The method shown is implemented.

[0097] Based on the same concept as the intelligent alarm triggering device in the embodiment, the present invention also provides an intelligent alarm triggering method including:

[0098] Acquire a building structure design drawing of a target building, and extract multiple stairwells of the target building based on the building structure design drawing, wherein the multiple stairwells have multiple positioning building location identifiers, and the multiple stairwells include multiple floor fire door sets;

[0099] Retrieving multiple monitoring video sets of multiple camera sets of the fire door sets on the multiple floors within a preset monitoring period, identifying the states of the fire doors on the multiple monitoring video sets, and obtaining multiple first safety factor sets;

[0100] Collecting floor status information and floor evacuation partition information set of the target building within a preset monitoring period;

[0101] Performing environmental safety identification on multiple floor fire door sets of the target building according to the floor status information and the floor evacuation partition information set to obtain multiple second safety factor sets;

[0102] Performing fire safety level matching on the plurality of stairwells based on the plurality of first safety factor sets and the plurality of second safety factor sets to obtain a plurality of fire safety levels;

[0103] Based on the multiple fire safety levels, early warning is performed on the sensor monitoring data of the multiple stairwells to generate early warning instructions.

[0104] In some embodiments, the method comprises:

[0105] Constructing a state recognition network layer, wherein the state recognition network layer includes a first subject recognition branch, a first obstacle recognition branch, and a safety factor recognition branch;

[0106] The state recognition network layer is used to perform image analysis on the multiple monitoring video sets respectively to obtain the multiple first safety factor sets.

[0107] In some embodiments, the method comprises:

[0108] Collecting multiple sample images of fire doors on multiple sample floors at different opening and closing angles, wherein the multiple sample images have multiple angle identifiers;

[0109] Using a convolutional network layer to perform supervised learning on the mapping relationship between the multiple sample images and the multiple angle identifiers, and updating the network parameters according to the output results until the output reaches convergence, thereby obtaining the first subject recognition branch that has been trained;

[0110] Using the first subject recognition branch to dynamically recognize video frames in the plurality of surveillance video sets, respectively, and output a plurality of angle change sequence sets;

[0111] The multiple angle change sequence sets are respectively used as multiple first identification data sets, and are associated and mapped with multiple second identification data sets obtained after the first obstacle identification branch performs obstacle identification on the multiple surveillance video sets, to obtain multiple safety identification data sets;

[0112] The plurality of safety identification data sets are analyzed in the safety factor identification branch to obtain the plurality of first safety factor sets.

[0113] In some embodiments, the method comprises:

[0114] Collecting a plurality of sample angle change sequences, a plurality of sample second identification data and a plurality of sample first safety factors as construction data;

[0115] The constructed data is used as a plurality of analysis points in a two-dimensional analysis space, wherein a one-dimensional coordinate axis of the two-dimensional analysis space is a sample angle variation sequence, a two-dimensional coordinate axis is second identification data, and each analysis point has a sample first safety factor identifier;

[0116] The two-dimensional analysis space is embedded in the safety factor identification branch, and the multiple safety identification data sets are analyzed to obtain the multiple first safety factor sets.

[0117] In some embodiments, the method comprises:

[0118] Extracting the first security identification data from the plurality of security identification data sets, inputting the first security identification data into the two-dimensional analysis space, and obtaining a first point to be identified;

[0119] Calculate the Euclidean distances from a plurality of analysis points in the two-dimensional analysis space to the first point to be identified, and arrange the calculation results in ascending order, and use the first m analysis points as the target point set;

[0120] Performing weighted calculation on the first safety factor identifiers of the m samples of the target point set to generate a first safety factor of the first point to be identified;

[0121] The multiple safety identification data sets are analyzed using the two-dimensional analysis space to obtain the multiple first safety factor sets.

[0122] In some embodiments, the method further comprises:

[0123] Counting the pedestrian flow data of multiple floors of the target building within a preset monitoring period to obtain the pedestrian flow data of multiple floors;

[0124] respectively comparing the plurality of floor traffic data with the sum of the plurality of floor traffic data to obtain a plurality of floor density coefficients, and using the plurality of floor density coefficients as floor status information;

[0125] Extracting the evacuation guidance signs of the target building, and determining multiple evacuation planning area sets of multiple floor fire door sets;

[0126] Based on the multiple evacuation planning area sets, the floor evacuation zone information set is generated, and each floor evacuation zone information corresponds to a floor of the target building.

[0127] In some embodiments, the method further comprises:

[0128] Allocating the multiple floor density coefficient sets according to the floor evacuation partition information set to generate multiple fire door evacuation pressure coefficient sets corresponding to the multiple floor fire door sets;

[0129] The fire door evacuation pressure coefficients in the plurality of fire door evacuation pressure coefficient sets are respectively divided by the sum of the coefficients of the plurality of fire door evacuation pressure coefficient sets to obtain a plurality of second safety factor sets.

[0130] Embodiment 3

[0131] Based on the same concept as an intelligent alarm trigger device in the embodiment, the present invention also provides a fire door, which is an execution terminal of an intelligent alarm trigger device of the present invention. The fire door is provided with an electric controller. When the intelligent alarm trigger device sends an alarm signal, the electric controller is used to intelligently control the fire door.

[0132] It should be understood that the embodiments mentioned in this specification focus on their differences from other embodiments. The specific embodiments in the aforementioned embodiment one are also applicable to an intelligent alarm triggering method described in embodiment two. For the sake of brevity of the specification, they will not be further elaborated here.

[0133] It should be understood that the embodiments disclosed in the present invention and the above description can enable those skilled in the art to use the present invention to implement the present invention. At the same time, the present invention is not limited to the above-mentioned embodiments. It should be understood that those skilled in the art can still modify the technical solutions recorded in the above-mentioned embodiments, or replace some of the technical features therein by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. An intelligent alarm triggering device, characterized in that: The device comprises: A stairwell positioning module, the stairwell positioning module is used to obtain a building structure design drawing of a target building, and extract multiple stairwells of the target building based on the building structure design drawing, wherein the multiple stairwells have multiple positioning building location identifiers, and the multiple stairwells include multiple floor fire door sets; a door state recognition module, the door state recognition module being used to retrieve a plurality of monitoring video sets of a plurality of camera sets of the plurality of floor fire door sets within a preset monitoring period, perform fire door state recognition on the plurality of monitoring video sets, and obtain a plurality of first safety factor sets; A floor status information collection module, the floor status information collection module is used to collect floor status information and floor evacuation partition information set of the target building within a preset monitoring period; An environmental safety identification module, the environmental safety identification module is used to perform environmental safety identification on a plurality of floor fire door sets of the target building according to the floor status information and the floor evacuation partition information set, and obtain a plurality of second safety factor sets; A safety level matching module, the safety level matching module is used to perform fire safety level matching on the plurality of stairwells based on the plurality of first safety factor sets and the plurality of second safety factor sets to obtain a plurality of fire safety levels; An early warning module, the early warning module is used to issue an early warning to the sensor monitoring data of the plurality of stairwells based on the plurality of fire safety levels and generate an early warning instruction; The execution steps of the floor status information acquisition module in the device include: Counting the pedestrian flow data of multiple floors of the target building within a preset monitoring period to obtain the pedestrian flow data of multiple floors; respectively comparing the plurality of floor traffic data with the sum of the plurality of floor traffic data to obtain a plurality of floor density coefficients, and using the plurality of floor density coefficients as floor status information; Extracting the evacuation guidance signs of the target building, and determining multiple evacuation planning area sets of multiple floor fire door sets; Based on the multiple evacuation planning area sets, generating the floor evacuation partition information set, each floor evacuation partition information corresponds to a floor of the target building; The execution step of the environmental safety identification module in the device also includes: Allocating the multiple floor density coefficient sets according to the floor evacuation partition information set to generate multiple fire door evacuation pressure coefficient sets corresponding to the multiple floor fire door sets; The fire door evacuation pressure coefficients in the plurality of fire door evacuation pressure coefficient sets are respectively divided by the sum of the coefficients of the plurality of fire door evacuation pressure coefficient sets to obtain a plurality of second safety factor sets.

2. The device according to claim 1, characterized in that The execution steps of the door state identification module in the device include: Constructing a state recognition network layer, wherein the state recognition network layer includes a first subject recognition branch, a first obstacle recognition branch, and a safety factor recognition branch; The state recognition network layer is used to perform image analysis on the multiple monitoring video sets respectively to obtain the multiple first safety factor sets.

3. The device according to claim 2, characterized in that The execution steps of the door status identification module in the device also include: Collecting multiple sample images of fire doors on multiple sample floors at different opening and closing angles, wherein the multiple sample images have multiple angle identifiers; Using a convolutional network layer to perform supervised learning on the mapping relationship between the multiple sample images and the multiple angle identifiers, and updating the network parameters according to the output results until the output reaches convergence, thereby obtaining the first subject recognition branch that has been trained; Using the first subject recognition branch to dynamically recognize video frames in the plurality of surveillance video sets, respectively, and output a plurality of angle change sequence sets; The multiple angle change sequence sets are respectively used as multiple first identification data sets, and are associated and mapped with multiple second identification data sets obtained after the first obstacle identification branch performs obstacle identification on the multiple surveillance video sets, to obtain multiple safety identification data sets; The plurality of safety identification data sets are analyzed in the safety factor identification branch to obtain the plurality of first safety factor sets.

4. The device according to claim 3, characterized in that The execution steps of the door status identification module in the device also include: Collecting a plurality of sample angle change sequences, a plurality of sample second identification data and a plurality of sample first safety factors as construction data; The constructed data is used as a plurality of analysis points in a two-dimensional analysis space, wherein a one-dimensional coordinate axis of the two-dimensional analysis space is a sample angle variation sequence, a two-dimensional coordinate axis is second identification data, and each analysis point has a sample first safety factor identifier; The two-dimensional analysis space is embedded in the safety factor identification branch, and the multiple safety identification data sets are analyzed to obtain the multiple first safety factor sets.

5. The device according to claim 4, characterized in that The execution steps of the door status identification module in the device also include: Extracting the first security identification data from the plurality of security identification data sets, inputting the first security identification data into the two-dimensional analysis space, and obtaining a first point to be identified; Calculate the Euclidean distances from a plurality of analysis points in the two-dimensional analysis space to the first point to be identified, and arrange the calculation results in ascending order, and use the first m analysis points as the target point set; Performing weighted calculation on the first safety factor identifiers of the m samples of the target point set to generate a first safety factor of the first point to be identified; The multiple safety identification data sets are analyzed using the two-dimensional analysis space to obtain the multiple first safety factor sets.

6. An intelligent alarm triggering method, characterized in that: The method comprises: Acquire a building structure design drawing of a target building, and extract multiple stairwells of the target building based on the building structure design drawing, wherein the multiple stairwells have multiple positioning building location identifiers, and the multiple stairwells include multiple floor fire door sets; Retrieving multiple monitoring video sets of multiple camera sets of the fire door sets on the multiple floors within a preset monitoring period, identifying the states of the fire doors on the multiple monitoring video sets, and obtaining multiple first safety factor sets; Collecting floor status information and floor evacuation partition information set of the target building within a preset monitoring period; Performing environmental safety identification on multiple floor fire door sets of the target building according to the floor status information and the floor evacuation partition information set to obtain multiple second safety factor sets; Performing fire safety level matching on the plurality of stairwells based on the plurality of first safety factor sets and the plurality of second safety factor sets to obtain a plurality of fire safety levels; Based on the multiple fire safety levels, early warning is performed on the sensor monitoring data of the multiple stairwells to generate early warning instructions; The method further comprises: Counting the pedestrian flow data of multiple floors of the target building within a preset monitoring period to obtain the pedestrian flow data of multiple floors; respectively comparing the plurality of floor traffic data with the sum of the plurality of floor traffic data to obtain a plurality of floor density coefficients, and using the plurality of floor density coefficients as floor status information; Extracting the evacuation guidance signs of the target building, and determining multiple evacuation planning area sets of multiple floor fire door sets; Based on the multiple evacuation planning area sets, generating the floor evacuation partition information set, each floor evacuation partition information corresponds to a floor of the target building; The method further comprises: Allocating the multiple floor density coefficient sets according to the floor evacuation partition information set to generate multiple fire door evacuation pressure coefficient sets corresponding to the multiple floor fire door sets; The fire door evacuation pressure coefficients in the plurality of fire door evacuation pressure coefficient sets are respectively divided by the sum of the coefficients of the plurality of fire door evacuation pressure coefficient sets to obtain a plurality of second safety factor sets.

Citation Information

Patent Citations

  • Fireproof door with early warning and prompting functions

    CN106223812A

  • Controlling system and method for fireproof door of utility tunnel

    CN107575139A

  • Fireproof door control system and method, terminal and storage medium

    CN114233139A

  • Method for providing disaster optimal evacuation route information and system therefor

    WO2023014191A1

  • KR20230129137A