Fire alarm method and device based on spatio-temporal neighborhood correlation of multidimensional sensors

Abstract

The application relates to a fire alarm method and device based on a multidimensional sensor of space-time proximity association, and belongs to the technical field of intelligent fire alarm data processing. The method comprises the following steps: when any smoke alarm triggers a fire alarm signal, searching for online temperature sensors, humidity sensors and video monitoring devices within a target distance range and a target time range; determining a temperature change rate factor based on the temperature collected by the online temperature sensors, determining an environmental humidity factor based on the environmental humidity collected by the online humidity sensors, and determining a video AI recognition factor based on the monitoring area picture collected by the online video monitoring device; when it is determined that a fire occurs based on at least two of the temperature change rate factor, the environmental humidity factor and the video AI recognition factor, the determination is true fire alarm, and alarm pushing is performed. The fire alarm method provided by the application verifies alarm authenticity through the fusion of four types of dimensional data based on space-time association, and reduces the false alarm rate.

Description

Technical Field

[0001] This invention relates to the field of intelligent fire alarm data processing technology, and in particular to a fire alarm method and device based on a multi-dimensional sensor with spatiotemporal proximity correlation. Background Technology

[0002] Currently, intelligent fire protection systems are widely used in various civil and industrial buildings. Most buildings are equipped with a variety of fire sensing devices, such as smoke detectors, temperature sensors, humidity sensors, heat detectors, and video surveillance, forming a comprehensive three-dimensional monitoring network. The mainstream operating mode of existing fire alarm systems is independent monitoring and reporting by individual devices, primarily relying on the single trigger signal from a smoke sensor as the basis for fire alarm determination. This method is simple to deploy, has a fast response time, and can meet basic fire early warning needs, and is widely used in various scenarios such as residential communities, commercial office buildings, shopping malls, and factories. With the continuous improvement of fire protection intelligence requirements, the industry is gradually introducing auxiliary means such as video AI recognition and environmental parameter monitoring to compensate for the shortcomings of single-sensor monitoring and promote the upgrade of fire alarm determination methods from a single dimension to a multi-dimensional comprehensive determination.

[0003] Traditional fire alarm systems suffer from several technical defects in practical use, severely impacting system stability and maintenance efficiency. First, traditional solutions rely solely on smoke sensors for alarm detection, resulting in poor fault tolerance. Non-fire conditions such as high humidity, dust accumulation, slight temperature fluctuations, and equipment aging and drift can trigger false alarms, leading to a high false alarm rate. Frequent invalid alarms not only waste firefighting resources but also disrupt normal on-site maintenance. Second, the independent operation of monitoring devices and lack of data linkage prevent comprehensive verification using multi-dimensional information such as ambient temperature, humidity, and video footage. This makes it difficult to effectively distinguish between genuine fire signals and false signals caused by environmental interference, resulting in low alarm accuracy. Third, existing systems lack self-optimization capabilities. For recurring false alarms at fixed locations and time periods, the monitoring thresholds cannot be adjusted, leading to repeated occurrences of similar false alarms and failing to address the root cause. Fourth, the same real fire incident often triggers multiple and different types of monitoring devices to report alarms simultaneously. The monitoring platform will receive and display multiple duplicate alarm data, resulting in information redundancy, which greatly increases the workload of on-duty personnel in information identification and handling, and reduces the efficiency of fire emergency response.

[0004] Currently, while some improvement solutions exist in the industry for addressing fire false alarms and duplicate alarms, their optimization effects are limited, and significant technical shortcomings remain. Regarding false alarms, existing methods primarily employ fixed threshold filtering and manual masking of faulty equipment. These methods can only filter some abnormal signals based on fixed values, failing to combine the spatiotemporal correlation characteristics of equipment with multi-dimensional environmental data for dynamic comprehensive judgment. This results in low filtering accuracy, poor scenario adaptability, and an inability to meet monitoring needs across different time periods and environments. Furthermore, existing solutions lack historical data learning capabilities and cannot optimize equipment parameters based on long-term false alarm patterns, leading to the continued occurrence of similar false alarms. Regarding duplicate alarms, most existing systems employ simple time-based deduplication logic, only removing completely identical duplicate information. This fails to identify different types of related alarms from different devices arising from the same physical event, resulting in poor deduplication and persistent alarm redundancy. Summary of the Invention

[0005] In view of this, it is necessary to provide a fire alarm method and device based on multi-dimensional sensors with spatiotemporal proximity correlation to solve the problem of high false alarm rate of existing fire alarms.

[0006] To address the aforementioned problems, in a first aspect, the present invention provides a fire alarm method based on a multi-dimensional sensor with spatiotemporal proximity correlation, comprising: When any smoke detector triggers a fire alarm signal, search for online temperature sensors, humidity sensors, and video surveillance equipment within the target distance and target time range; The temperature change rate factor is determined based on the temperature collected by the online temperature sensor, the environmental humidity factor is determined based on the environmental humidity collected by the online humidity sensor, and the video AI recognition factor is determined based on the monitoring area images collected by the online video surveillance equipment. When a fire is determined to have occurred based on at least two of the temperature change rate factor, the ambient humidity factor, and the video AI recognition factor, it is identified as a real fire alarm and an alarm is pushed out.

[0007] One possible implementation also includes: When it is determined that no fire has occurred based on at least two of the temperature change rate factor, the ambient humidity factor, and the video AI recognition factor, it is judged as a false fire alarm.

[0008] One possible implementation also includes: When the number of false fire alarms from a smoke detector at the same location and during the same time period reaches a set threshold, the sensitivity of the smoke detector at that location and during that time period is reduced.

[0009] In one possible implementation, determining the temperature change rate factor based on the real-time temperature acquired by the online temperature sensor includes: Based on the real-time temperature, determine the rate of temperature change; When the temperature change rate is greater than the preset fire temperature rise threshold, the temperature change rate factor is assigned to 1. When the temperature change rate is less than or equal to the preset fire temperature rise threshold, the temperature change rate factor is assigned to 0.

[0010] In one possible implementation, determining the environmental humidity factor based on the environmental humidity collected by the online humidity sensor includes: When the ambient humidity is within the normal fire monitoring range, the ambient humidity factor is assigned a value of 1; When the ambient humidity is not within the normal fire monitoring range, the ambient humidity factor is assigned a value of 0.

[0011] In one possible implementation, determining the video AI recognition factor based on the monitoring area images collected by the online video surveillance equipment includes: The video image of the monitored area is detected using a video intelligent analysis algorithm; When flames or smoke are detected, the video AI recognition factor is assigned a value of 1; When no flames or smoke are detected, the video AI recognition factor is assigned to 0.

[0012] In one possible implementation, the target time range is the 10 seconds before and the 10 seconds after the fire alarm signal is triggered.

[0013] In one possible implementation, the target distance range is a radius of 5 meters centered on the smoke detector.

[0014] Secondly, the present invention also provides a fire alarm device based on a multi-dimensional sensor with spatiotemporal proximity correlation, comprising: The search unit is used to search for online temperature sensors, humidity sensors and video surveillance equipment within the target distance range and target time range when any smoke detector triggers a fire alarm signal. The determining unit is used to determine the temperature change rate factor based on the temperature collected by the online temperature sensor, to determine the environmental humidity factor based on the environmental humidity collected by the online humidity sensor, and to determine the video AI recognition factor based on the monitoring area image collected by the online video surveillance equipment. The alarm unit is used to determine a real fire alarm and send an alarm notification when at least two of the temperature change rate factor, the ambient humidity factor and the video AI recognition factor are used to determine that a fire has occurred.

[0015] Thirdly, the present invention also provides an electronic device, including a data acquisition unit, a memory, and a processor, wherein, The data acquisition unit, the memory, and the processor are communicatively connected. The data acquisition unit is used to receive fire alarm signals triggered by smoke detectors; The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors as described in any of the above implementations.

[0016] The beneficial effects of this invention are as follows: The fire alarm method and device based on spatiotemporal proximity correlation of this invention can search for online temperature sensors, humidity sensors and video surveillance equipment within the target distance range and target time range when any smoke detector triggers a fire alarm signal. This allows for linkage with online monitoring equipment around the alarm smoke detector and ensures the temporal consistency of monitoring data from multiple devices. If a fire is determined to have occurred based on at least two of the temperature change rate factor, environmental humidity factor and video AI recognition factor, the alarm is considered a real fire alarm and an alarm push is executed. Otherwise, it is determined to be a false alarm. This invention abandons the traditional single-device, single-parameter judgment mode and comprehensively verifies the authenticity of the alarm by fusing four types of data through spatiotemporal correlation. This can effectively filter invalid false alarms and reduce the false alarm rate of fire alarms. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic flowchart of one embodiment of the fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors provided by the present invention; Figure 2 This is a second schematic flowchart of an embodiment of the fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors provided by the present invention; Figure 3 The overall flowchart of the multi-dimensional sensor collaborative false alarm filtering and alarm merging method based on spatiotemporal proximity correlation provided by the present invention; Figure 4 A schematic diagram of an embodiment of the fire alarm device based on spatiotemporal proximity correlation of the present invention; Figure 5 A schematic diagram of an embodiment of the electronic device provided by the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0020] In the description of the embodiments of the present invention, unless otherwise stated, "multiple" means two or more. "And / or" describes the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.

[0021] The terms "first," "second," etc., used in the embodiments of this invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a technical feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature.

[0022] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0023] This invention provides a fire alarm method and device based on a multi-dimensional sensor with spatiotemporal proximity correlation, which will be described below.

[0024] Figure 1 This is one of the schematic flowcharts of an embodiment of the fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors provided by the present invention, as shown below. Figure 1 As shown, the fire alarm method based on multi-dimensional sensors with spatiotemporal proximity correlation includes: S101. When any smoke detector triggers a fire alarm signal, search for online temperature sensors, humidity sensors and video surveillance equipment within the target distance range and target time range. S102. Determine the temperature change rate factor based on the temperature collected by the online temperature sensor, determine the environmental humidity factor based on the environmental humidity collected by the online humidity sensor, and determine the video AI recognition factor based on the monitoring area image collected by the online video surveillance equipment. S103. When a fire is determined to have occurred based on at least two of the temperature change rate factor, the ambient humidity factor, and the video AI recognition factor, it is determined to be a real fire alarm, and an alarm is pushed.

[0025] It should be noted that this invention collects monitoring data from all smoke detectors, temperature sensors, humidity sensors, and video surveillance equipment in real time, 24 / 7, and monitors the operating status of each device in real time.

[0026] In S101, when any smoke detector triggers an alarm signal, the system starts a spatiotemporal proximity retrieval mechanism to dynamically match surrounding compliant monitoring devices within the target distance range and target time range, and form a temporary multi-dimensional sensor collaborative monitoring group (i.e., online temperature sensors, humidity sensors and video surveillance equipment).

[0027] In some embodiments of the present invention, the target time range is the 10 seconds before and the 10 seconds after the fire alarm signal is triggered.

[0028] In some embodiments of the present invention, the target distance range is a range with a radius of 5 meters centered on the smoke detector.

[0029] This invention pre-configures a spatial proximity threshold S of 5 meters, and only triggers online monitoring devices within a 5-meter radius of the alarm smoke detector.

[0030] The present invention pre-configures the time synchronization threshold T to 10 seconds to ensure the time consistency of monitoring data from multiple devices.

[0031] For example, when any smoke detector alarm i At any moment t When an alarm signal is triggered, all online monitoring devices on site are traversed, and those that simultaneously satisfy the conditions of spatial distance D(i,j) ≤ S and data acquisition time within [ tT , t+T Temperature sensors, humidity sensors, and video surveillance equipment within the specified range automatically form a dynamic sensor collaborative group with smoke detectors that trigger alarms as the core.

[0032] This collaborative group does not require pre-defined device binding relationships and can be dynamically matched in real time according to alarm points to adapt to different monitoring scenarios.

[0033] In S102, this invention constructs a four-dimensional fire detection factor and determines the authenticity of the alarm through a multi-parameter comprehensive verification method. By reading the temperature change rate and ambient humidity data of the equipment within the collaborative group, and simultaneously retrieving the video AI recognition results, the authenticity of the alarm is verified through a multi-dimensional joint judgment model.

[0034] For example, based on the real-time smoke concentration data collected by the smoke detector, when the smoke concentration data reaches the basic alarm threshold of the device, the smoke detector signal factor A=1 is assigned; otherwise, A=0 is assigned.

[0035] Calculate the real-time temperature change rate ΔT / Δt of the temperature sensor within the collaborative group. When the temperature change rate is greater than the preset fire temperature rise threshold, assign a temperature change rate factor B=1; otherwise, assign B=0.

[0036] Collect real-time monitoring data from humidity sensors within the collaborative group to identify interference from high humidity environments. When the ambient humidity is within the normal fire monitoring range, assign an ambient humidity factor C=1; otherwise, assign C=0.

[0037] The video intelligent analysis algorithm detects the images in the monitored area. When fire features such as flames and smoke are identified, the video AI recognition factor D=1 is assigned; otherwise, D=0 is assigned.

[0038] In S103, under the premise of smoke detection signal triggering (A=1), if at least two of the three auxiliary factors, namely temperature change rate factor, ambient humidity factor and video AI recognition factor, are determined to be valid fire characteristics, the alarm is identified as a real fire alarm; otherwise, it is determined to be a false alarm.

[0039] If the alarm is determined to be a genuine fire alarm, the system will perform alarm push and event archiving operations.

[0040] If a false alarm is detected, the system retains complete data including the location, time period, and environmental parameters of the false alarm. It also continuously counts the cumulative number of false alarms at each location and during each fixed time period. When the number reaches a preset threshold, the system automatically lowers the sensitivity threshold of the smoke detectors at the corresponding location and during the corresponding time period.

[0041] This embodiment also screens all alarm data on the monitoring platform in real time, performs merging and deduplication processing on multiple alarms with the same source and related information, and generates standardized event records.

[0042] By repeatedly executing the above monitoring, judgment, optimization, and merging processes, the system can achieve continuous dynamic operation and autonomous optimization.

[0043] In summary, the fire alarm method based on spatiotemporal proximity correlation of multi-dimensional sensors provided in this invention searches for online temperature sensors, humidity sensors, and video surveillance equipment within the target distance and time range when any smoke detector triggers a fire alarm signal. This links online monitoring equipment around the alarm smoke detector and ensures the temporal consistency of monitoring data from multiple devices. If a fire is determined to have occurred based on at least two of the temperature change rate factor, environmental humidity factor, and video AI recognition factor, the alarm is considered a genuine fire alarm, and an alarm push is executed. Otherwise, it is determined to be a false alarm. This invention abandons the traditional single-device, single-parameter judgment mode and comprehensively verifies the authenticity of the alarm by fusing four types of data through spatiotemporal correlation, which can effectively filter invalid false alarms and reduce the false alarm rate of fire alarms.

[0044] In some embodiments of the present invention, it further includes: When it is determined that no fire has occurred based on at least two of the temperature change rate factor, the ambient humidity factor, and the video AI recognition factor, it is judged as a false fire alarm.

[0045] In some embodiments of the present invention, it further includes: When the number of false fire alarms from a smoke detector at the same location and during the same time period reaches a set threshold, the sensitivity of the smoke detector at that location and during that time period is reduced.

[0046] This embodiment establishes a dedicated false alarm statistics ledger, which fully records core data such as device number, monitoring point, time period, and on-site environmental parameters for each false alarm event. A cumulative false alarm threshold N (e.g., 3 times) is pre-set, and historical false alarm data for the same monitoring point and the same fixed time period are cumulatively statistically analyzed.

[0047] When the cumulative number of false alarms at a certain location and during a certain time period reaches or exceeds the set threshold N, the system will automatically lower the sensitivity threshold of the smoke detector at that location by one level during the corresponding time period, thereby improving the device's fault tolerance to normal environmental interference and preventing the recurrence of similar false alarms.

[0048] The sensitivity of smoke detectors is divided into 5 levels. Each optimization only lowers the sensitivity by 1 level to avoid excessive threshold adjustment.

[0049] This embodiment adopts a gradient-level adjustment method, with a fixed adjustment range for each threshold, which can effectively avoid the impact of sudden threshold changes on the accuracy of normal fire monitoring.

[0050] This embodiment also sets up a matching rule for alarms originating from the same source: for multiple alarm messages uploaded by different monitoring devices and channels within the same spatial location and the same time window (±T seconds), they are uniformly identified as related alarms derived from the same physical event. The system automatically extracts key features such as spatial location, occurrence time, monitoring area, and event type of multiple alarms for correlation matching, integrating and merging multiple duplicate and scattered alarm data into a complete standardized event record. The merged event record integrates all valid monitoring data, removes redundant and duplicate information, and pushes only a single standardized record to the monitoring platform.

[0051] In some embodiments of the present invention, determining the temperature change rate factor based on the real-time temperature collected by the online temperature sensor includes: Based on the real-time temperature, determine the rate of temperature change; When the temperature change rate is greater than the preset fire temperature rise threshold, the temperature change rate factor is assigned to 1. When the temperature change rate is less than or equal to the preset fire temperature rise threshold, the temperature change rate factor is assigned to 0.

[0052] Calculate the real-time temperature change rate ΔT / Δt based on the real-time temperature from the temperature sensors within the collaborative group.

[0053] When the rate of temperature change is greater than the preset fire temperature rise threshold, the rate of temperature change factor B is assigned to 1; otherwise, B is assigned to 0.

[0054] The preset fire temperature rise threshold is set to 2℃ / 10s to quickly identify fire temperature rise characteristics.

[0055] In some embodiments of the present invention, determining the environmental humidity factor based on the environmental humidity collected by the online humidity sensor includes: When the ambient humidity is within the normal fire monitoring range, the ambient humidity factor is assigned a value of 1; When the ambient humidity is not within the normal fire monitoring range, the ambient humidity factor is assigned a value of 0.

[0056] By collecting real-time monitoring data from humidity sensors within the collaborative group, high humidity environmental interference can be identified. When the ambient humidity is within the normal fire monitoring range, the ambient humidity factor C=1 is assigned; otherwise, C=0 is assigned.

[0057] In some embodiments of the present invention, determining the video AI recognition factor based on the monitoring area images collected by the online video surveillance equipment includes: The video image of the monitored area is detected using a video intelligent analysis algorithm; When flames or smoke are detected, the video AI recognition factor is assigned a value of 1; When no flames or smoke are detected, the video AI recognition factor is assigned to 0.

[0058] The video intelligent analysis algorithm detects the images in the monitored area. When fire features such as flames and smoke are identified, the video AI recognition factor D=1 is assigned; otherwise, D=0 is assigned.

[0059] Video intelligent analysis algorithms can be based on single-frame image target detection algorithms such as YOLO or Faster R-CNN.

[0060] The joint judgment rule in this embodiment is as follows: under the premise that the smoke detector signal is triggered (A=1), if at least two of the three auxiliary factors, namely the temperature change rate, the ambient humidity, and the video AI recognition, are determined to be valid fire characteristics, the alarm is considered to be a real fire alarm; otherwise, it is considered to be a false alarm.

[0061] This invention centers on smoke detectors that trigger alarms. Based on the principles of matching device spatial distance and data acquisition time, it dynamically constructs a multi-sensor collaborative monitoring group. Combining smoke concentration signals, temperature change rates, ambient humidity data, and video AI recognition results, it builds a multi-dimensional judgment model to accurately distinguish between real fire alarms and false alarms. Simultaneously, this solution automatically records and statistically analyzes historical false alarm data for each location and time period. In high-frequency false alarm scenarios, it adaptively lowers the monitoring sensitivity of smoke detectors, reducing repeated false alarms at the source. Furthermore, through alarm spatiotemporal correlation matching logic, it merges repeated alarms from multiple channels and devices generated by the same event into a standardized event record, eliminating platform data redundancy and achieving precise, adaptive, and intelligent processing of fire alarm monitoring.

[0062] Figure 2 This is a second schematic flowchart of an embodiment of the fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors provided by the present invention, as shown below. Figure 2 As shown, it includes: Step S201: Preliminary parameter configuration.

[0063] The system pre-configures basic operating parameters, and all parameters can be flexibly adjusted according to building type, equipment performance, and site environment. The default configuration in this embodiment is as follows: 1. The spatial proximity threshold S is set to 5 meters, which will only trigger the online monitoring devices within a 5-meter radius of the smoke detector. 2. The time synchronization threshold T is set to 10 seconds to ensure the time consistency of monitoring data from multiple devices; 3. The cumulative false alarm threshold N is set to 3 times. The threshold optimization mechanism will be triggered if there are 3 false alarms at the same location and in the same period. 4. The sensitivity of the smoke detector is divided into 5 levels. Each optimization will only lower the sensitivity by 1 level to avoid excessive threshold adjustment. 5. The fire temperature change threshold is set to 2℃ / 10s to quickly identify fire temperature rise characteristics.

[0064] Step S202: Implementation of the dynamic sensor collaborative group.

[0065] The system continuously polls the operating status and real-time data of all monitoring devices 24 hours a day. When a smoke detector triggers an alarm signal, the system immediately searches for all online devices within a 5-meter radius and 10 seconds before and after the device's spatial coordinates. It successfully detects nearby temperature sensors, humidity sensors, and high-definition video surveillance equipment that are normally online, and automatically adds these devices to the dynamic collaborative monitoring group corresponding to this alarm, completing the temporary linkage configuration.

[0066] Step S203: Implementation of multi-dimensional data joint alarm determination.

[0067] After the collaborative group is established, the system synchronously collects monitoring data from each device and completes factor determination: the smoke detector detects that the smoke concentration reaches the basic alarm threshold, factor A=1; the temperature sensor in the collaborative group heats up by 0.8℃ within 10 seconds, which does not reach the 2℃ fire threshold, factor B=0; the real-time ambient humidity is 78%, which is a high humidity interference environment, factor C=0; no flame or smoke characteristics are detected in the video monitoring screen, factor D=0.

[0068] According to the joint judgment rules, if only one factor is valid, the alarm is judged as a false alarm. If a real fire occurs on site, with a rapid rise in temperature and fire characteristics identified by video, and two or more auxiliary factors are valid, the system can determine it as a real fire alarm, immediately push alarm information, retain a complete set of monitoring data, and trigger the emergency early warning mechanism.

[0069] Step S204: False alarm data recording and adaptive threshold optimization implementation.

[0070] The system fully archives false alarm events, recording information including device number, monitoring location, time of false alarm, ambient humidity, ambient temperature, video monitoring footage, and device operating status.

[0071] The system performs cumulative counting based on the statistical dimension of "location + time period". If the number of false alarms at a location reaches 3 within a monitoring period on a given day, the system automatically lowers the sensitivity threshold of the smoke detector at that location by one level each day for that time period. This improves the device's fault tolerance to interference from high humidity environments and prevents repeated false alarms during that period. The device parameters remain unchanged for other time periods, ensuring no impact on the accuracy of normal fire monitoring.

[0072] Step S205: Implementation of alarm association merging optimization.

[0073] When a real fire occurs in the area, the smoke detector, temperature sensor, and heat detector will trigger an alarm sequentially within a 10-second time window.

[0074] The system uses spatiotemporal correlation rules to determine that multiple alarm messages originating from the same spatial location and time interval are related alarms derived from the same physical fire. The system automatically integrates valid data from multiple alarms, including smoke concentration, temperature changes, environmental parameters, and video footage, eliminating duplicate and redundant alarm entries. Finally, it merges these into a standardized fire event record containing comprehensive monitoring information and pushes it to the monitoring platform. This effectively avoids the accumulation of numerous duplicate alarms and provides on-duty personnel with concise, complete, and accurate event information.

[0075] Step S206: Implement the full-process cyclical operation.

[0076] After completing the single alarm determination, false alarm recording, parameter optimization, and alarm merging process, the system continuously maintains real-time monitoring of all equipment and cyclically executes all workflows. Simultaneously, it continuously updates the false alarm statistics log, dynamically iterates and optimizes the equipment monitoring thresholds at each location and time period, and continuously improves the system's alarm determination accuracy, achieving automated, intelligent, and adaptive operation throughout the entire fire monitoring process.

[0077] Figure 3 The flowchart of the multi-dimensional sensor collaborative false alarm filtering and alarm merging method based on spatiotemporal proximity correlation provided by the present invention fully demonstrates the entire process logic of the present invention from real-time data acquisition, smoke alarm triggering, dynamic collaborative group formation, multi-dimensional data joint judgment, false alarm data statistics and threshold adaptive optimization, same-source alarm merging to final event output. It clearly reflects the connection relationship of each technical module, the branch judgment logic and the loop operation mechanism, and clearly presents the overall technical architecture of the present invention.

[0078] The fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors provided by this invention has the following advantages: (1) Effectively reduce the false alarm rate of the system and improve the accuracy of alarm judgment. This invention abandons the traditional judgment mode of single device and single parameter, and dynamically builds a collaborative monitoring group through spatiotemporal correlation. It integrates four types of data to comprehensively verify the authenticity of the alarm, which can accurately distinguish between environmental interference, equipment failure and real fire, and greatly filter out invalid false alarms. It fundamentally solves the problem of single judgment logic and poor fault tolerance of traditional solutions.

[0079] (2) It has the ability to learn and optimize data and eliminate repeated false alarms. This invention can automatically collect and count historical false alarm data, accurately locate the locations and time periods of high-frequency false alarms, and adapt to the interference characteristics of the local environment by adaptively lowering the equipment sensitivity threshold. It can suppress the recurrence of similar false alarms from the source without the need for manual and repeated debugging of equipment parameters, thus effectively reducing the workload of equipment operation and maintenance.

[0080] (3) Eliminate alarm data redundancy and improve the efficiency of on-duty operation and maintenance. This invention can accurately identify multiple devices and multiple channels related alarms derived from the same fire event, automatically complete the merging and deduplication, and output only a single standardized event record, which completely solves the problem of alarm stacking and repeated display on the platform, reduces the workload of on-duty personnel in information identification and repeated handling, and improves the efficiency of fire emergency response.

[0081] (4) Strong equipment adaptability and high resource utilization. The present invention adopts a dynamic group formation mode, which does not require pre-fixed equipment linkage relationship. It can match the surrounding effective monitoring equipment in real time according to the alarm location and time, and can adapt to various complex building layouts and equipment deployment scenarios. It avoids resource idleness and waste caused by fixed linkage, and greatly improves the system's operational flexibility and adaptability.

[0082] (5) Fully automated operation with a high degree of intelligence. From data acquisition, alarm judgment, false alarm statistics, threshold optimization to alarm merging output, the entire process of this invention does not require manual intervention. It can autonomously complete iterative optimization and data processing, reducing human error and meeting the development needs of unmanned operation and intelligent maintenance in smart fire protection.

[0083] To better implement the fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors in the embodiments of the present invention, based on the fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors, correspondingly, as follows: Figure 4 As shown, this embodiment of the invention also provides a fire alarm device based on a multi-dimensional sensor with spatiotemporal proximity correlation. The fire alarm device 400 based on a multi-dimensional sensor with spatiotemporal proximity correlation includes: The search unit 401 is used to search for online temperature sensors, humidity sensors and video surveillance equipment within the target distance range and target time range when any smoke detector triggers a fire alarm signal. The determining unit 402 is used to determine the temperature change rate factor based on the temperature collected by the online temperature sensor, to determine the environmental humidity factor based on the environmental humidity collected by the online humidity sensor, and to determine the video AI recognition factor based on the monitoring area image collected by the online video surveillance equipment. The alarm unit 403 is used to determine a real fire alarm and send an alarm push when a fire is determined to have occurred based on at least two of the temperature change rate factor, the ambient humidity factor and the video AI recognition factor.

[0084] The fire alarm device 400 based on spatiotemporal proximity correlation of the multi-dimensional sensor provided in the above embodiments can realize the technical solution described in the above embodiments of the fire alarm method based on spatiotemporal proximity correlation of the multi-dimensional sensor. The specific implementation principle of each module or unit can be found in the corresponding content in the above embodiments of the fire alarm method based on spatiotemporal proximity correlation of the multi-dimensional sensor, which will not be repeated here.

[0085] like Figure 5 As shown, the present invention also provides an electronic device 500. The electronic device 500 includes a processor 501, a memory 502, a display 503, and a data acquisition device 504. Figure 5 Only some components of the electronic device 500 are shown, but it should be understood that it is not required to implement all the components shown, and more or fewer components may be implemented instead.

[0086] In some embodiments, processor 501 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 502 or process data, such as the fire alarm method based on spatiotemporal proximity correlation of multi-dimensional sensors in this invention.

[0087] In some embodiments, processor 501 may be a single server or a group of servers. The server group may be centralized or distributed. In some embodiments, processor 501 may be local or remote. In some embodiments, processor 501 may be implemented on a cloud platform. In some embodiments, the cloud platform may include a private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, internal cloud, multi-cloud, or any combination thereof.

[0088] In some embodiments, memory 502 may be an internal storage unit of electronic device 500, such as a hard disk or memory of electronic device 500. In other embodiments, memory 502 may also be an external storage device of electronic device 500, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on electronic device 500.

[0089] Furthermore, the memory 502 may include both internal storage units of the electronic device 500 and external storage devices. The memory 502 is used to store application software and various types of data installed on the electronic device 500.

[0090] In some embodiments, display 503 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an organic light-emitting diode (OLED) touchscreen, etc. Display 503 is used to display information from electronic device 500 and to display a visual user interface.

[0091] In some embodiments, the data acquisition unit 504 may be a sensor for receiving fire alarm signals triggered by a smoke detector.

[0092] Components 501-504 of electronic device 500 communicate with each other via a system bus.

[0093] In one embodiment, when processor 501 executes a fire alarm program based on a multi-dimensional sensor with spatiotemporal proximity correlation stored in memory 502, the following steps can be implemented: When any smoke detector triggers a fire alarm signal, search for online temperature sensors, humidity sensors, and video surveillance equipment within the target distance and target time range; The temperature change rate factor is determined based on the temperature collected by the online temperature sensor, the environmental humidity factor is determined based on the environmental humidity collected by the online humidity sensor, and the video AI recognition factor is determined based on the monitoring area images collected by the online video surveillance equipment. When a fire is determined to have occurred based on at least two of the temperature change rate factor, the ambient humidity factor, and the video AI recognition factor, it is identified as a real fire alarm and an alarm is pushed out.

[0094] It should be understood that when the processor 501 executes the fire alarm program based on the spatiotemporal proximity correlation of the multi-dimensional sensor in the memory 502, in addition to the functions mentioned above, it can also perform other functions, as can be found in the description of the corresponding method embodiments above.

[0095] Furthermore, this embodiment of the invention does not specifically limit the type of electronic device 500 mentioned. Electronic device 500 can be a mobile phone, tablet computer, personal digital assistant (PDA), wearable device, laptop computer, or other portable electronic device. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices running iOS, Android, Microsoft, or other operating systems. The aforementioned portable electronic device can also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the invention, electronic device 500 may not be a portable electronic device, but rather a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0096] Accordingly, embodiments of the present invention also provide a computer-readable storage medium for storing computer-readable programs or instructions. When the programs or instructions are executed by a processor, they can implement the steps or functions of the fire alarm method based on spatiotemporal proximity correlation of multi-dimensional sensors provided in the above-described method embodiments.

[0097] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.), and the computer program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0098] The fire alarm method and device based on spatiotemporal proximity correlation multidimensional sensors provided by the present invention have been described in detail above. Specific examples have been used to illustrate the principle and implementation of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A fire alarm method based on a multi-dimensional sensor with spatiotemporal proximity correlation, characterized in that, include: When any smoke detector triggers a fire alarm signal, search for online temperature sensors, humidity sensors, and video surveillance equipment within the target distance and target time range; The temperature change rate factor is determined based on the temperature collected by the online temperature sensor, the environmental humidity factor is determined based on the environmental humidity collected by the online humidity sensor, and the video AI recognition factor is determined based on the monitoring area images collected by the online video surveillance equipment. When a fire is determined to have occurred based on at least two of the temperature change rate factor, the ambient humidity factor, and the video AI recognition factor, it is identified as a real fire alarm and an alarm is pushed out.

2. The fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors according to claim 1, characterized in that, Also includes: When it is determined that no fire has occurred based on at least two of the temperature change rate factor, the ambient humidity factor, and the video AI recognition factor, it is judged as a false fire alarm.

3. The fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors according to claim 2, characterized in that, Also includes: When the number of false fire alarms from a smoke detector at the same location and during the same time period reaches a set threshold, the sensitivity of the smoke detector at that location and during that time period is reduced.

4. The fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors according to claim 1, characterized in that, The determination of the temperature change rate factor based on the real-time temperature collected by the online temperature sensor includes: Based on the real-time temperature, determine the rate of temperature change; When the temperature change rate is greater than the preset fire temperature rise threshold, the temperature change rate factor is assigned to 1. When the temperature change rate is less than or equal to the preset fire temperature rise threshold, the temperature change rate factor is assigned to 0.

5. The fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors according to claim 1, characterized in that, The determination of the environmental humidity factor based on the environmental humidity collected by the online humidity sensor includes: When the ambient humidity is within the normal fire monitoring range, the ambient humidity factor is assigned a value of 1; When the ambient humidity is not within the normal fire monitoring range, the ambient humidity factor is assigned a value of 0.

6. The fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors according to claim 1, characterized in that, The process of determining video AI recognition factors based on the monitoring area images collected by the online video surveillance equipment includes: The video image of the monitored area is detected using a video intelligent analysis algorithm; When flames or smoke are detected, the video AI recognition factor is assigned a value of 1; When no flames or smoke are detected, the video AI recognition factor is assigned to 0.

7. The fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors according to claim 1, characterized in that, The target time range is the 10 seconds before and the 10 seconds after the fire alarm signal is triggered.

8. The fire alarm method based on spatiotemporal proximity correlation of multidimensional sensors according to claim 1, characterized in that, The target distance range is defined as a radius of 5 meters centered on the smoke detector.

9. A fire alarm device based on a multi-dimensional sensor with spatiotemporal proximity correlation, characterized in that, include: The search unit is used to search for online temperature sensors, humidity sensors and video surveillance equipment within the target distance range and target time range when any smoke detector triggers a fire alarm signal. The determining unit is used to determine the temperature change rate factor based on the temperature collected by the online temperature sensor, to determine the environmental humidity factor based on the environmental humidity collected by the online humidity sensor, and to determine the video AI recognition factor based on the monitoring area image collected by the online video surveillance equipment. The alarm unit is used to determine a real fire alarm and send an alarm notification when a fire is determined to have occurred based on at least two of the temperature change rate factor, the ambient humidity factor, and the video AI recognition factor.

10. An electronic device, characterized in that, Includes a data acquisition unit, memory, and processor, among which, The data acquisition unit, the memory, and the processor are communicatively connected. The data acquisition unit is used to receive fire alarm signals triggered by smoke detectors; The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the fire alarm method based on spatiotemporal proximity correlation of a multidimensional sensor as described in any one of claims 1 to 8.

Images