Fire detection method and apparatus, electronic device, and storage medium

Abstract

The application provides a fire detection method and device, an electronic device and a storage medium. The method comprises the following steps: acquiring construction information based on an interface connection with a construction management platform; generating an interference factor based on the construction information, the interference factor being used to represent the interference degree of fire detection; acquiring sensor data for fire detection from a detection terminal; and determining a fire detection result based on the sensor data and the interference factor. In this way, the accuracy of fire detection can be improved in a high-floor construction scene.

Description

Technical Field

[0001] This application relates to the field of fire detection technology, and in particular to a fire detection method, device, electronic equipment and storage medium. Background Technology

[0002] Currently, for high-rise construction scenarios (such as renovations), the construction process can interfere with fire detection (e.g., dust interference, welding interference, etc.). Existing practices typically reduce false alarms by fusing multi-sensor data, such as using a combination of smoke, temperature, and carbon monoxide sensors. However, simple multi-sensor data fusion analysis is insufficient to handle complex construction scenarios (e.g., during welding operations, high-temperature particles and dust are present simultaneously, making the aforementioned combination of smoke and temperature sensors prone to failure). Therefore, improving the accuracy of fire detection in high-rise construction scenarios is a pressing technical problem that needs to be solved in this field. Summary of the Invention

[0003] This application provides a fire detection method, device, electronic equipment, and storage medium, which can improve the accuracy of fire detection in high-rise construction scenarios.

[0004] In a first aspect, this application provides a fire detection method applied to an edge computing gateway in a fire detection system, which also includes a detection terminal and an alarm. The method includes:

[0005] Construction information is obtained through an interface connection with the construction management platform;

[0006] Interference factors are generated based on construction information, and these factors are used to characterize the degree of interference with fire detection.

[0007] Acquire sensor data for fire detection from the detection terminal;

[0008] Fire detection results are determined based on sensor data and interference factors.

[0009] As can be seen, in this application, the construction management platform provides construction activity data, the edge computing gateway generates interference factors based on construction information as dynamic weight parameters to adjust the sensor data processing logic, and the edge computing gateway fuses the interference factors and sensor data locally in real time to generate fire detection results. This process quantifies interference factors through construction information, enabling the system to distinguish between real fire and construction interference, thereby improving the accuracy of fire detection.

[0010] In a feasible example, the construction information includes a list of work plans for the current floor and adjacent floors above and below. Each work plan list includes one or more work plans, and each work plan includes a work type, three-dimensional coordinate location, planned start time, and planned end time. When the work plan list includes only one work plan, interference factors are generated based on the construction information, including:

[0011] The weight of a task type is determined based on a preset mapping table and the task type. The preset mapping table includes the correspondence between task types and task type weights. The higher the probability that a task type will cause a false fire alarm, the higher the task type weight.

[0012] The spatial distance weight is determined based on the distance between the three-dimensional coordinate position and the position of the detection terminal; the smaller the distance, the greater the spatial distance weight.

[0013] Based on the inclusion relationship between the first time interval and the first time, the weight of the time window is determined. The first time interval is determined based on the planned start time and the planned end time, and the first time is the time when construction information is obtained.

[0014] Interference factors are generated based on job type weight, spatial distance weight, and time window weight.

[0015] In this application, the work plan list of the current floor and adjacent floors above and below is obtained through construction information. The weight of the work type is determined by a preset mapping table, and the weight of spatial distance and time window are calculated. These weights are fused to generate an interference factor, which is used to adjust the sensor data processing logic. This can achieve the technical effect of distinguishing between real fire and construction interference and improving the alarm accuracy.

[0016] In a feasible example, when the work schedule list includes multiple work schedules, interference factors are generated based on construction information, including:

[0017] Obtain the first interference factor for each of the multiple work plans. The first interference factor is determined based on the work type, three-dimensional coordinate position, start time and end time of each work plan.

[0018] Interference factors are generated based on multiple first interference factors.

[0019] In this application, by aggregating multiple first interference factors to generate a final interference factor, the spatiotemporal superposition effect between operations can be taken into account. At this time, the edge computing gateway can locally fuse the interference factors and sensor data in real time, dynamically adjust the sensitivity to filter construction interference, distinguish real fire situations, and improve alarm accuracy.

[0020] In a feasible example, interference factors are generated based on multiple first interference factors, including:

[0021] The first operation value is obtained by performing a cube operation on each of the multiple first interference factors and then summing the results.

[0022] The second operational value is obtained by summing multiple first interference factors;

[0023] The interference factor is obtained by taking the square root of the ratio between the first and second operands.

[0024] In this application, the first calculated value can be the sum of the cubes of each first interference factor. Performing cube operations on each of the multiple first interference factors and then summing the results can be done by iterating through each first interference factor, performing the cube operations, and then summing the results, thereby generating a numerical result that enhances the contribution of high-risk operations. The second calculated value can be the numerical result of directly summing multiple first interference factors, and can be used as a normalized weight.

[0025] In a feasible example, fire detection results are determined based on sensor data and interference factors, including:

[0026] Feature vectors are constructed based on sensor data and interference factors;

[0027] The feature vector is input into the classification model to obtain the probabilities corresponding to the first classification result, the second classification result, and the third classification result, respectively. The first classification result is used to represent no fire, the second classification result is used to represent suspected fire, and the third classification result is used to represent confirmed fire.

[0028] The fire detection results are determined based on the probabilities corresponding to the first, second, and third classification results, respectively.

[0029] In this application, by learning the characteristic differences between construction interference and actual fire through a classification model, and determining the fire detection result based on probability values, dynamic compensation based on interference factors can be achieved, avoiding false alarms based solely on sensor threshold judgments. At the same time, by providing decision confidence through probability quantification, the technical effect of improving the alarm accuracy in complex construction scenarios can be achieved.

[0030] In a feasible example, the fire detection result is determined based on the probabilities corresponding to the first classification result, the second classification result, and the third classification result, respectively, including:

[0031] If the probability of responding to the third classification result is greater than or equal to the first threshold, then the fire detection result is determined to be a confirmed fire.

[0032] If the probability of the third classification result is less than the first threshold and the probability of the second classification result is greater than or equal to the second threshold, then the fire detection result is determined to be a suspected fire.

[0033] If the probability of the third classification result is less than the first threshold, the probability of the second classification result is less than the second threshold, and the probability of the first classification result is greater than or equal to the third threshold, then the fire detection result is determined to be no fire.

[0034] If the probability of the third classification result is less than the first threshold, the probability of the second classification result is less than the second threshold, and the probability of the first classification result is less than the third threshold, then the fire detection result is determined to be a suspected fire.

[0035] In this application, a multi-level threshold judgment is performed based on the probabilities of the first, second, and third classification results. When the probability of the third classification result exceeds the first threshold, a real fire is confirmed. When the probability of the third classification result is insufficient but the probability of the second classification result exceeds the second threshold, a suspected fire is marked. When the probability of the first classification result exceeds the third threshold, it is determined to be normal construction dust. If all probabilities are below the threshold, it is classified as a suspected fire. This can achieve fine-grained risk classification, accurately distinguish between real fires, construction interference, and normal construction activities, and significantly improve the alarm accuracy in complex construction scenarios.

[0036] In a feasible example, the edge computing gateway connects to a cloud server, and the fire detection system also includes an alarm. The method further includes:

[0037] If the fire detection result confirms a fire, the alarm will be activated to sound a fire alarm.

[0038] If the fire detection result indicates a suspected fire, a review request is sent to the cloud server. The review request is used to instruct the review of the fire detection result.

[0039] In this application, the alarm is triggered locally when a fire is confirmed, and a verification request is sent to the cloud server when a fire is suspected. This mechanism achieves the synergy between rapid edge response and accurate cloud verification, which not only ensures the immediate handling of real fires, but also significantly reduces the false alarm rate in construction scenarios and improves the reliability of the system in complex high-rise environments.

[0040] Secondly, this application provides a fire detection device applied to an edge computing gateway in a fire detection system. The fire detection system also includes a detection terminal. The device includes:

[0041] The communication unit is used to obtain construction information through an interface connection with the construction management platform;

[0042] The processing unit is used to generate interference factors based on construction information. The interference factors are used to characterize the degree of interference to fire detection.

[0043] The communication unit is also used to acquire sensor data for fire detection from the detection terminal;

[0044] The processing unit is also used to determine the fire detection results based on sensor data and interference factors.

[0045] Thirdly, this application provides an electronic device including a processor, a memory, and a communication interface. The processor, memory, and communication interface are interconnected and perform communication with each other. The memory stores executable program code, the communication interface is used for wireless communication, and the processor is used to retrieve the executable program code stored in the memory and execute some or all of the steps described in any of the methods in the first aspect.

[0046] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements some or all of the steps described in the first aspect of this application.

[0047] Fifthly, this application provides a computer program product, including a computer program that, when processed and executed, implements some or all of the steps described in the first aspect of this application. The computer program product may be a software installation package. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a schematic diagram of the structure of a fire detection system provided in an embodiment of this application;

[0050] Figure 2 A schematic flowchart illustrating a fire detection method provided in an embodiment of this application;

[0051] Figure 3 A flowchart illustrating another fire detection method provided in this application embodiment;

[0052] Figure 4 A schematic flowchart illustrating yet another fire detection method provided in this application embodiment;

[0053] Figure 5 A functional unit block diagram of a fire detection device provided in an embodiment of this application;

[0054] Figure 6 A functional unit block diagram of another fire detection device provided in the embodiments of this application;

[0055] Figure 7 This is a structural block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0056] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0057] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps is not limited to the steps listed, but may optionally include steps not listed, or may optionally include other steps inherent to these processes, methods, products, or apparatuses.

[0058] 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 this application. 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.

[0059] Currently, in high-rise construction scenarios, the construction process can interfere with fire detection (e.g., dust interference, welding interference, cutting interference, etc.). Typical technologies for handling construction scenarios employ a combination of smoke sensors, temperature sensors, and carbon monoxide sensors. The core logic is: when the smoke concentration exceeds a preset threshold A, and the temperature rise rate exceeds a preset threshold B, or the carbon monoxide concentration exceeds a preset threshold C, a fire is detected and an alarm is triggered. Some products introduce simple AND logic (e.g., alarm only when both smoke and carbon monoxide exceed the limits simultaneously) to reduce false alarms. However, in complex construction scenarios, such as welding operations, where high-temperature particles and dust are present simultaneously, the judgment of smoke concentration and temperature rise rate may fail. Furthermore, some technologies deploy high-definition network cameras at key locations on the construction site, running smoke and flame recognition algorithms on a backend video analysis server. These algorithms are typically based on color models, motion features, and texture analysis to detect smoke areas or flickering flames in the video footage. When a suspected fire is detected, the system displays an alarm screen at the monitoring center. However, during construction, the dense scaffolding and protective netting can easily obstruct the camera's field of view, and video analysis requires frame-by-frame processing, which consumes a lot of computing resources.

[0060] Based on this, this solution provides a fire detection method. It acquires relevant construction information through a construction platform, generates corresponding interference factors based on this information, and finally analyzes the fire detection results by combining the interference factors with data from other sensors. This approach comprehensively considers the impact of construction on fire detection, improving the accuracy of fire detection in high-rise construction scenarios.

[0061] Please see Figure 1 , Figure 1 This is a schematic diagram of the structure of a fire detection system provided in an embodiment of this application, as shown below. Figure 1 As shown, the fire detection system 100 includes an edge computing gateway 101, a detection terminal 102, an alarm 103, and a cloud server 104.

[0062] The edge computing gateway 101 is a computing unit deployed on a construction floor or within a construction area. It is responsible for aggregating data from multiple detection terminals 102 in that area, executing local decision-making algorithms, and synchronizing data with the cloud server 104. In practical applications, one gateway can be deployed on each floor. Its core can be an industrial-grade processor (e.g., a quad-core ARM Cortex-A53, 1.5GHz), running an embedded system, and featuring a built-in lightweight classification model engine (e.g., a lightweight gradient booster machine, LightGBM). It also possesses LoRa / Zigbee aggregation, Ethernet / 4G communication, local storage, and GPS / BeiDou timing capabilities. Furthermore, the edge computing gateway 101 can connect to the construction management platform via a construction information interface to obtain relevant construction information.

[0063] The detection terminal 102 is deployed in key locations on each construction floor (such as stairwells, elevator shafts, and material storage areas). It includes at least: a laser scattering dust sensor (PM2.5 / PM10, range 0-1000 μg / m³), an electrochemical carbon monoxide sensor (range 0-500 ppm), a thermopile temperature sensor (range -40~125℃), a microcontroller, and a Lora / Zigbee wireless communication module. The terminal has a built-in real-time clock and supports local timestamp marking.

[0064] The alarm 103 includes an audible and visual alarm (DC24V driven), a fire broadcast (IP network broadcast module), an SMS / APP notification module (via a 4G module), and an optional low-resolution thermal imaging camera (for remote verification in case of suspected fire). When the edge computing gateway 101 confirms a fire, it will control the alarm 103 to sound an alarm.

[0065] The cloud server 104 operates the management system, time-series database, and web visualization platform. After the edge computing gateway 101 sends a manual review request to the cloud server 104, the cloud server 104 can also send the manual review request to the terminal device corresponding to the security officer.

[0066] In this application, the edge computing gateway 101 connects to the construction management platform through an interface to obtain construction information. Then, based on the construction information, it generates an interference factor to characterize the degree of interference to fire detection. Next, it obtains sensor data for fire detection from the detection terminal 102. Finally, based on the sensor data and the interference factor, it determines the fire detection result and controls the alarm 103 to sound an alarm based on the fire detection result.

[0067] Based on this, the present application provides a fire detection method, and the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0068] Example 1: The main process of the fire detection method is described below.

[0069] Please see Figure 2 , Figure 2 This is a flowchart illustrating a fire detection method provided in an embodiment of this application. The method is applied to the aforementioned edge computing gateway, such as... Figure 2 As shown, the method includes the following steps.

[0070] Step S201: Obtain construction information based on the interface connection with the construction management platform.

[0071] The construction management platform can be a software system for managing construction activities (such as Glodon BIM 5D or Luban Workshop), and can be used to provide real-time data on construction activities. In this embodiment, the construction management platform can connect to an edge computing gateway via an API interface to transmit construction information. Furthermore, obtaining construction information can be achieved by calling the construction management platform interface via a REST API to obtain JSON-formatted construction information, or by subscribing to construction events published by the construction management platform through a message queue, thereby obtaining real-time data on construction activities. The construction information can be detailed data describing the construction activities and can be used to generate interference factors to characterize the degree of interference.

[0072] Step S202: Generate interference factors based on construction information.

[0073] The interference factor can be a parameter that numerically represents the impact of construction interference on fire detection. It can be used to adjust the weight of sensor data and reduce false alarms. In this embodiment, the interference factor is calculated based on construction information, such as welding activities triggering high interference factor values. Furthermore, the interference factor can be generated by matching a preset interference value table with the activity type in the construction information and dynamically calculating the interference weight based on the time and location information in the construction information, thereby outputting a numerical parameter representing the degree of interference.

[0074] Step S203: Obtain sensor data for fire detection from the detection terminal.

[0075] The detection terminal senses environmental changes through physical sensor elements and outputs electrical signals. Furthermore, sensor data, including smoke concentration, temperature, and carbon monoxide concentration, is output through the sensor elements of the detection terminal. Moreover, sensor data can be acquired by reading the real-time data stream from the detection terminal via serial communication protocol and by receiving sensor data packets sent by the detection terminal via wireless protocol, thereby obtaining the raw input data for fire detection.

[0076] Step S204: Determine the fire detection results based on sensor data and interference factors.

[0077] The fire detection result can be a fire status judgment based on sensor data and interference factors, which can be used to determine whether to trigger an alarm and what type of alarm to trigger. Determining the fire detection result can be achieved by combining interference factors and sensor data to calculate the fire status. Furthermore, determining the fire detection result can be accomplished by adjusting the sensor data threshold as a weight using the interference factor, and using a machine learning model to fuse the interference factor and sensor data to output probability values, thereby outputting the fire status judgment result.

[0078] In this embodiment, the construction management platform provides construction activity data, and the edge computing gateway generates interference factors based on construction information as dynamic weight parameters to adjust the sensor data processing logic. The edge computing gateway fuses the interference factors and sensor data locally in real time to generate fire detection results. This process quantifies interference factors through construction information, enabling the system to distinguish between real fires and construction interference, thereby improving the accuracy of fire detection.

[0079] In Example 2, the construction information includes a list of work plans for the current floor and adjacent floors above and below. Each work plan includes one or more work plans, with each plan specifying its type, three-dimensional coordinates, start time, and end time. The current floor and adjacent floors above and below can represent the entire floor range involved in the construction activity, which can be used to limit the scope of construction information collection and optimize interference calculations. In this example, the current floor and adjacent floors above and below can be obtained from the construction management platform and represented by a list of floor numbers. The work plan list can be a structured data set containing multiple work plans, providing detailed planning information for construction activities.

[0080] The following section provides a detailed explanation of the fire detection method, assuming the work plan list contains only one work plan, based on the details of determining the interference factor.

[0081] Please see Figure 3 , Figure 3 This is a flowchart illustrating another fire detection method provided in an embodiment of this application. This method is applied to the aforementioned edge computing gateway, such as... Figure 3 As shown, the method includes the following steps.

[0082] Step S301: Obtain construction information based on the interface connection with the construction management platform.

[0083] Step S302: Determine the job type weight based on the preset mapping table and job type.

[0084] The preset mapping table includes the correspondence between job types and job type weights. The higher the probability that a job type will cause a false fire alarm, the higher the job type weight. The job type can be a classification identifier that identifies the nature of the construction activity. In this embodiment, the job type can be extracted from the work plan as a string or an enumeration value. The job type can include one or more of the following: welding job type, cutting job type, grinding job type, dust job type, etc.

[0085] The preset mapping table can be configuration data that stores the correspondence between job types and weights, and can be used to provide the basis for mapping job type weights. In this embodiment, the preset mapping table is stored locally on the edge computing gateway and can be queried in key-value pair form. The job type weight can be a numerical risk coefficient calculated based on the job type, which can be used to characterize the probability of a job type causing a false alarm. For example, welding job = 0.9, cutting job = 0.6, grinding job = 0.5, pouring job = 0.3, no job = 0.

[0086] Step S303: Determine the spatial distance weight based on the distance between the three-dimensional coordinate position and the position of the detection terminal.

[0087] In this context, the smaller the distance, the greater the spatial distance weight. The three-dimensional coordinate position can be coordinate data describing the spatial location of construction activities, and can be used to calculate the spatial distance weight and quantify the impact of location. In this embodiment, the three-dimensional coordinate position can be obtained from the work plan and represented in a three-dimensional coordinate system. For example, the three-dimensional coordinate position can include one or more of absolute coordinate positions, relative coordinate positions, and floor coordinate positions.

[0088] The spatial distance weight can be a numerical influence coefficient calculated based on the location distance, which can be used to characterize the interference intensity of the construction location on the detection terminal. In this embodiment, the spatial distance weight is calculated by the Euclidean distance between the three-dimensional coordinate position and the detection terminal position, and inversely mapped to the weight. For example, a Gaussian decay function can be used for weight mapping. The formula is:

[0089] .

[0090] in, Used to represent spatial distance weights The exponential function is used to represent the distance between the three-dimensional coordinate position and the position of the detection terminal, and σ is the spatial influence scale parameter (e.g., preset to 10 meters, which can be adjusted according to the size of the construction site).

[0091] Step S304: Determine the weight of the time window based on the inclusion relationship between the first time interval and the first time.

[0092] The first time interval can be a time period defined by the planned start time and the planned end time, which can be used to determine whether the construction activity is within the current time's influence range. The first time can be the current time point when the construction information is obtained, which can be used for time window weight calculation to characterize real-time performance. In this embodiment, the first time is obtained from the system clock in timestamp format. The planned start time can be the starting time point of the work plan, and the planned end time can be the ending time point of the work plan. The planned start time and planned end time can be extracted from the work plan and stored in timestamp format.

[0093] In actual construction, the start time of work may generally be earlier and the end time may generally be later. Therefore, when defining the first time interval based on the planned start time and the planned end time, a corresponding threshold can be subtracted from the planned start time and the corresponding threshold can be added to the planned end time. That is, the first time interval is [T_start - 5 minutes, T_end + 5 minutes]. Where T_start is the planned start time and T_end is the planned end time.

[0094] The time window weight can be a numerical influence coefficient calculated based on time relationships, and can be used to characterize the persistence of the current disturbance caused by the construction time. In this embodiment, the time window weight determines whether the first time is within the first time interval and outputs a binary weight. If the first time is within the first time interval, the output time window weight is 1; if the first time is not within the first time interval, the output time window weight is 0.

[0095] Step S305: Generate interference factors based on job type weight, spatial distance weight, and time window weight.

[0096] The interference factor can be a dynamically adjusted coefficient that integrates multiple weighting parameters, and can be used to adjust the sensor data processing logic. In this embodiment, the interference factor is synthesized based on the job type weight, spatial distance weight, and time window weight, and is implemented through weighted averaging or product operations. For example, the interference factor can include one or more of high interference factor, medium interference factor, and low interference factor. The calculation formula for the interference factor is:

[0097] .

[0098] Where P represents the interference factor. Used to represent the weight of job type. Used to represent spatial distance weights Used to represent the weight of the time window.

[0099] In this embodiment, the work plan list of the current floor and adjacent floors above and below is obtained through construction information. The weight of the work type is determined by a preset mapping table, and the weight of spatial distance and time window are calculated. These weights are fused to generate an interference factor, which is used to adjust the sensor data processing logic. This can achieve the technical effect of distinguishing between real fire and construction interference and improving the alarm accuracy.

[0100] Step S306: Obtain sensor data for fire detection from the detection terminal;

[0101] Step S307: Determine the fire detection result based on sensor data and interference factors.

[0102] Example 3: When the work plan list includes multiple work plans, the fire detection method will be described in detail based on the determination details of the interference factor.

[0103] Please see Figure 4 , Figure 4 This is a flowchart illustrating another fire detection method provided in an embodiment of this application. This method is applied to the aforementioned edge computing gateway, such as... Figure 4 As shown, the method includes the following steps.

[0104] Step S401: Obtain construction information based on the interface connection with the construction management platform.

[0105] Step S402: Obtain the first interference factor for each of the multiple job plans.

[0106] The first interference factor is determined based on the job type, three-dimensional coordinate position, start time, and end time of each job plan.

[0107] Understandably, the calculation method for the first interference factor is similar to that in Example 2. Both methods determine the weight of the task type based on the task type, the weight of the spatial distance based on the distance between the three-dimensional coordinate position and the position of the detection terminal, and the weight of the time window based on the inclusion relationship between the time interval formed by the planned start time and the planned end time and the first time. The results are then calculated based on these three weight values. Further details will not be elaborated here.

[0108] Step S403: Generate interference factors based on multiple first interference factors.

[0109] The generation of interference factors based on multiple first interference factors can be achieved by aggregating these factors to obtain the final interference parameters. Furthermore, the generation of interference factors based on multiple first interference factors can be achieved by using a maximum value rule to select the highest interference factor, or by weighted summation of all first interference factors, thereby outputting a final parameter that comprehensively represents the interference from multiple tasks.

[0110] In this embodiment, by aggregating multiple first interference factors to generate a final interference factor, the spatiotemporal superposition effect between operations can be taken into account. At this time, the edge computing gateway locally fuses the interference factors and sensor data in real time, and can dynamically adjust the sensitivity to filter construction interference, distinguish real fire situations, and improve alarm accuracy.

[0111] In an optional embodiment, generating an interference factor based on a plurality of first interference factors includes: performing a cube operation on each of the plurality of first interference factors and then summing the results to obtain a first operation value; summing the plurality of first interference factors to obtain a second operation value; and performing a square root operation on the ratio between the first operation value and the second operation value to obtain the interference factor.

[0112] The first calculated value can be the sum of the cubed results of multiple first interference factors. It can be used to enhance the interference contribution of high interference operations, amplify high values ​​and compress low values, so that the strong interference signal almost "covers" the weak signal, making the fusion result closer to the real impact of the operation with the highest interference.

[0113] In this embodiment, the first calculated value can be the sum of the cubes of each first interference factor. Performing cube operations on each of the multiple first interference factors and then summing the results can be done by iterating through each first interference factor, performing the cube operations, and then summing the results, thereby generating a numerical result that enhances the contribution of high-risk operations. The second calculated value can be the numerical result of directly summing the multiple first interference factors, and can be used as a normalized weight.

[0114] For example, the formula for calculating the interference factor in this embodiment is as follows:

[0115] .

[0116] Where P represents the interference factor. The j-th first interference factor is used to represent the j-th first interference factor, and N is used to represent the number of first interference factors.

[0117] In this embodiment, the first operation value is obtained by summing the cubes of multiple first interference factors, the second operation value is obtained by directly summing the cubes, and the ratio is calculated and squared to generate interference factors. This can enhance the interference contribution of high-risk operations, balance the superposition effect of multiple operations, and perform nonlinear normalization processing. Moreover, this simple operation method can be executed quickly on the edge computing gateway, ensuring the technical effect of quickly and accurately distinguishing between real fire and construction interference in complex construction scenarios.

[0118] Step S404: Obtain sensor data for fire detection from the detection terminal.

[0119] Step S405: Determine the fire detection result based on sensor data and interference factors.

[0120] In an optional embodiment, determining the fire detection result based on sensor data and interference factors includes: constructing a feature vector based on sensor data and interference factors; inputting the feature vector into a classification model to obtain the probabilities corresponding to a first classification result, a second classification result, and a third classification result, respectively, wherein the first classification result is used to characterize no fire, the second classification result is used to characterize a suspected fire, and the third classification result is used to characterize a confirmed fire; and determining the fire detection result based on the probabilities corresponding to the first classification result, the second classification result, and the third classification result.

[0121] The feature vector can be a feature representation that fuses sensor data and interference factors, and can be used to provide standardized input features for the classification model. In this embodiment, the feature vector can be formed by concatenating sensor data and interference factors into a vector in a fixed order.

[0122] The classification model can be a machine learning model for classifying fire states, outputting a probability distribution of fire states based on input feature vectors. In an exemplary embodiment, the classification model is trained using historical fire and construction disturbance data, employing a supervised learning algorithm. The classification model can include convolutional neural network models, random forest models, distributed gradients based on decision tree algorithms (e.g., the LightGBM model), etc. Furthermore, the model can be iterated continuously at certain intervals, for example, a scheduled task is started every Monday at 00:00 to incrementally train the classification model using newly accumulated labeled data from the past week (as the training set) and historical data from the past month (as the validation set). The focus is on optimizing the classification boundary to adapt to changes in the construction phase (e.g., from the main construction phase to the decoration phase, changes in dust particle size distribution and composition).

[0123] The first classification result can represent a category indicating that no fire has occurred, such as being in a normal construction state or not under construction and also lacking fire risk. The second classification result can represent a category indicating a suspected fire state, which can be used to identify potential fire risks that require further verification. The second classification result is output by the classification model, and the corresponding probability value represents the likelihood of a suspected fire. The third classification result can represent a category indicating a confirmed fire state, which can be used to identify actual fire events. The third classification result is output by the classification model, and the corresponding probability value represents the likelihood of a fire occurring.

[0124] This probability can be a numerical confidence score output by the classification model, used to quantify the reliability of each classification result. This probability value can be calculated using the softmax function of the classification model.

[0125] Determining the fire detection result based on the probabilities corresponding to the first, second, and third classification results can be a way to judge the final fire status based on probability values. Furthermore, determining the fire detection result based on the probabilities corresponding to the first, second, and third classification results can involve comparing the probability values ​​and taking the category with the highest probability as the result, or setting a threshold to trigger an alarm when the probability of the third classification result exceeds the threshold, thereby outputting an accurate fire status judgment.

[0126] In this embodiment, by learning the characteristic differences between construction interference and actual fire through a classification model, and determining the fire detection result based on probability values, dynamic compensation based on interference factors can be achieved, avoiding false alarms based solely on sensor threshold judgments. At the same time, by providing decision confidence through probability quantification, the technical effect of improving the alarm accuracy in complex construction scenarios can be achieved.

[0127] Furthermore, in an optional embodiment, the fire detection result is determined based on the probabilities corresponding to the first classification result, the second classification result, and the third classification result, respectively, including: if the probability of the third classification result is greater than or equal to a first threshold, then the fire detection result is determined to be a confirmed fire; if the probability of the third classification result is less than the first threshold and the probability of the second classification result is greater than or equal to a second threshold, then the fire detection result is determined to be a suspected fire; if the probability of the third classification result is less than the first threshold, the probability of the second classification result is less than the second threshold, and the probability of the first classification result is greater than or equal to a third threshold, then the fire detection result is determined to be no fire; if the probability of the third classification result is less than the first threshold, the probability of the second classification result is less than the second threshold, and the probability of the first classification result is less than the third threshold, then the fire detection result is determined to be a suspected fire.

[0128] The first threshold can be the lowest probability threshold for confirming a fire, serving as a critical value to strictly distinguish between a real fire and a disturbance event. In this embodiment, the first threshold can be determined through training with historical data and is typically set to a high value to reduce false alarms. The second threshold can be the lowest probability threshold for identifying a suspected fire, used to mark potential risk events when the probability of a real fire is insufficient. In this embodiment, the second threshold can be set using comparative data of construction disturbances and actual fires, and is typically lower than the first threshold. The third threshold can be the lowest probability threshold for distinguishing between fires that have not occurred.

[0129] If the fire detection result is confirmed as a fire, the fire status can be output when the probability of the third classification result meets the threshold. The probability of the third classification result can be compared with the first threshold directly through numerical comparison calculation, or a conditional statement can be used to check whether the probability value is greater than or equal to the threshold.

[0130] Determining a fire detection result as a suspected fire can be done in two ways: either when the probability of the third category result is insufficient but the probability of the second category result meets the standard, or when the probability of the first, second, and third category results all fail to meet the standard. Furthermore, determining a fire detection result as a suspected fire can initiate both a cloud-based review process and a manual verification process.

[0131] Determining a fire detection result as "no fire" can be achieved when the probability of the third category result is below the threshold, the probability of the second category result is also below the threshold, but the probability of the first category result is within the threshold. The third category result is introduced because construction work can significantly impact fire detection; in such cases, the third category result can further determine whether a fire has occurred. Therefore, even when the first, second, and third category results are all below the threshold, a "suspected fire" status should still be output to avoid misinterpreting an early fire as a normal situation caused by construction work.

[0132] In this embodiment, a multi-level threshold judgment is performed based on the probabilities of the first, second, and third classification results. When the probability of the third classification result exceeds the first threshold, a real fire is confirmed. When the probability of the third classification result is insufficient but the probability of the second classification result exceeds the second threshold, a suspected fire is marked. When the probability of the first classification result exceeds the third threshold, it is determined to be normal construction dust. If all probabilities are below the threshold, it is classified as a suspected fire. This can achieve fine-grained risk classification, accurately distinguish between real fires, construction interference, and normal construction activities, and significantly improve the alarm accuracy in complex construction scenarios.

[0133] Furthermore, in an optional embodiment, the edge computing gateway is connected to a cloud server and controls the alarm to sound an alarm based on the fire detection results, including: in response to a confirmed fire, controlling the alarm to sound a fire alarm; in response to a suspected fire, sending a verification request to the cloud server, the verification request being used to instruct the verification of the fire detection results.

[0134] The cloud server can be a remotely deployed distributed computing resource, which can be used to provide secondary verification capabilities and reduce false positive rates. In this embodiment, the cloud server can receive verification requests through a network interface and perform data processing and analysis tasks.

[0135] A verification request can be instruction data used to trigger cloud verification, or it can be used to transmit fire detection results for secondary confirmation. In this embodiment, the verification request is generated by the edge computing gateway and includes details of the fire detection results and context information. For example, verification requests can include, but are not limited to, automatic verification requests, manual verification requests, and multi-source verification requests. Sending a verification request to the cloud server can involve transmitting the verification request data to the cloud server via the network. In the case of a manual verification request, the cloud server will also push the corresponding verification request to the safety officer's terminal device.

[0136] The alarm can be an alarm output device in a fire detection system, used to emit audible and visual alarm signals under trigger conditions. In this embodiment, the alarm receives control commands from the edge computing gateway to activate the alarm mechanism. Controlling the alarm to sound an alarm can be achieved by sending an alarm trigger command to the alarm. Furthermore, controlling the alarm to sound an alarm can be achieved by controlling the alarm power supply via GPIO signals or by sending an alarm trigger message via a network protocol.

[0137] In this embodiment, the alarm is triggered locally when a fire is confirmed, and a verification request is sent to the cloud server when a fire is suspected. This mechanism achieves the synergy between rapid edge response and accurate cloud verification, which not only ensures the immediate handling of real fires, but also significantly reduces the false alarm rate in construction scenarios and improves the reliability of the system in complex high-rise environments.

[0138] For embodiments consistent with those shown above, please refer to... Figure 5 , Figure 5 This is a functional unit block diagram of a fire detection device provided in an embodiment of this application. The fire detection device is a part of the aforementioned edge computing gateway. Figure 5 As shown, the fire detection device 50 includes:

[0139] The communication unit 501 is used to obtain construction information through an interface connection with the construction management platform;

[0140] Processing unit 502 is used to generate interference factors based on construction information. The interference factors are used to characterize the degree of interference to fire detection.

[0141] The communication unit 501 is also used to acquire sensor data for fire detection from the detection terminal;

[0142] The processing unit 502 is also used to determine the fire detection results based on sensor data and interference factors.

[0143] In one feasible embodiment, the construction information includes a list of work plans for the current floor and adjacent floors above and below. The work plan list includes one or more work plans, and each work plan includes a work type, three-dimensional coordinate position, planned start time, and planned end time. When the work plan list includes only one work plan, in generating interference factors based on the construction information, the processing unit 502 is specifically used for:

[0144] The weight of a task type is determined based on a preset mapping table and the task type. The preset mapping table includes the correspondence between task types and task type weights. The higher the probability that a task type will cause a false fire alarm, the higher the task type weight.

[0145] The spatial distance weight is determined based on the distance between the three-dimensional coordinate position and the position of the detection terminal; the smaller the distance, the greater the spatial distance weight.

[0146] Based on the inclusion relationship between the first time interval and the first time, the weight of the time window is determined. The first time interval is determined based on the planned start time and the planned end time, and the first time is the time when construction information is obtained.

[0147] Interference factors are generated based on job type weight, spatial distance weight, and time window weight.

[0148] In a feasible embodiment, when the work plan list includes multiple work plans, in terms of generating interference factors based on construction information, the processing unit 502 is specifically used for:

[0149] Obtain the first interference factor for each of the multiple work plans. The first interference factor is determined based on the work type, three-dimensional coordinate position, start time and end time of each work plan.

[0150] Interference factors are generated based on multiple first interference factors.

[0151] In a feasible embodiment, in terms of generating interference factors based on a plurality of first interference factors, the processing unit 502 is specifically configured to:

[0152] The first operation value is obtained by performing a cube operation on each of the multiple first interference factors and then summing the results.

[0153] The second operational value is obtained by summing multiple first interference factors;

[0154] The interference factor is obtained by taking the square root of the ratio between the first and second operands.

[0155] In one feasible embodiment, in determining the fire detection result based on sensor data and interference factors, the processing unit 502 is specifically used for:

[0156] Feature vectors are constructed based on sensor data and interference factors;

[0157] The feature vector is input into the classification model to obtain the probabilities corresponding to the first classification result, the second classification result, and the third classification result, respectively. The first classification result is used to represent no fire, the second classification result is used to represent suspected fire, and the third classification result is used to represent confirmed fire.

[0158] The fire detection results are determined based on the probabilities corresponding to the first, second, and third classification results, respectively.

[0159] In a feasible embodiment, in determining the fire detection result based on the probabilities corresponding to the first classification result, the second classification result, and the third classification result, the processing unit 502 is specifically used for:

[0160] If the probability of responding to the third classification result is greater than or equal to the first threshold, then the fire detection result is determined to be a confirmed fire.

[0161] If the probability of the third classification result is less than the first threshold and the probability of the second classification result is greater than or equal to the second threshold, then the fire detection result is determined to be a suspected fire.

[0162] If the probability of the third classification result is less than the first threshold, the probability of the second classification result is less than the second threshold, and the probability of the first classification result is greater than or equal to the third threshold, then the fire detection result is determined to be no fire.

[0163] If the probability of the third classification result is less than the first threshold, the probability of the second classification result is less than the second threshold, and the probability of the first classification result is less than the third threshold, then the fire detection result is determined to be a suspected fire.

[0164] In one feasible embodiment, the communication unit 501 is further configured to:

[0165] If the fire detection result confirms a fire, the alarm will be activated to sound a fire alarm.

[0166] If the fire detection result indicates a suspected fire, a review request is sent to the cloud server. The review request is used to instruct the review of the fire detection result.

[0167] It is understood that since the method embodiments and the device embodiments are different presentations of the same technical concept, the content of the method embodiment section in this application should be adapted to the device embodiment section in a synchronous manner, and will not be repeated here.

[0168] When using integrated units, such as Figure 6 As shown, Figure 6 This is a block diagram illustrating the functional unit composition of another fire detection device provided in an embodiment of this application. Figure 6 In this document, the fire detection device 50 includes a processing module 612 and a communication module 611. The processing module 612 controls and manages the operation of the fire detection device 50, for example, the steps of the processing unit 502, and / or other processes for executing the techniques described herein. The communication module 611 supports interaction between the fire detection device 50 and other devices, for example, the steps of the communication unit 501. Figure 6 As shown, the fire detection device 50 may also include a storage module 613, which is used to store the program code and data of the fire detection device 50.

[0169] The processing module 612 can be a processor or controller, such as a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an ASIC, an FPGA, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc. The communication module 611 can be a transceiver, RF circuitry, or a communication interface, etc. The storage module 613 can be a memory.

[0170] All relevant content for each scenario involved in the above method embodiments can be referenced from the functional descriptions of the corresponding functional modules, and will not be repeated here. All of the above fire detection devices 50 can execute the fire detection methods shown in Embodiments 1 to 3.

[0171] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0172] Figure 7 This is a structural block diagram of an electronic device provided in an embodiment of this application. Figure 7 As shown, the electronic device 700 may include one or more of the following components: processor 701, memory 702 and communication interface 703. The processor 701, memory 702 and communication interface 703 are interconnected and perform communication between them. The memory 702 may store one or more computer programs. The one or more computer programs may be configured to implement the methods described in the above embodiments when executed by one or more processors 701.

[0173] Processor 701 may include one or more processing cores. Processor 701 connects to various parts within the electronic device 700 using various interfaces and lines, and performs various functions and processes data of the electronic device 700 by running or executing instructions, programs, code sets, or instruction sets stored in memory 702, and by calling data stored in memory 702. Optionally, processor 701 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). Processor 701 may integrate one or more of a Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. It is understood that the aforementioned modem may also not be integrated into processor 701, but may be implemented separately through a communication chip.

[0174] The memory 702 may include random access memory (RAM) or read-only memory (ROM). The memory 702 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 702 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as touch functionality, sound playback functionality, image playback functionality, etc.), and instructions for implementing the various method embodiments described above. The data storage area may also store data created by the electronic device 700 during use.

[0175] It is understood that the electronic device 700 may include more or fewer structural elements than those shown in the above block diagram, such as a power module, physical buttons, WiFi (Wireless Fidelity) module, speaker, Bluetooth module, sensor, etc., without limitation.

[0176] The aforementioned electronic device 700 may be an edge computing gateway or a part of an edge computing gateway.

[0177] This application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements some or all of the steps of any of the fire detection methods described in the above method embodiments.

[0178] This application also provides a computer program product, including a computer program that, when executed by a processor, implements some or all of the steps of any of the fire detection methods described in the above method embodiments. This computer program product can be a software installation package.

[0179] It should be noted that, for the sake of simplicity, all embodiments of the aforementioned fire detection methods are described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to this application.

[0180] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple instances. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce a good effect.

[0181] Those skilled in the art will understand that all or part of the steps in the various method embodiments of any of the above fire detection methods can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage device, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a disk, or an optical disk, etc.

[0182] The embodiments of this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of a fire detection method, device, electronic device, and storage medium of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and its core ideas of this application. At the same time, for those skilled in the art, based on the ideas of a fire detection method, device, electronic device, and storage medium of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

[0183] This application is described with reference to flowchart illustrations and / or block diagrams of methods, hardware products, and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0184] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0185] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0186] It is understood that any product that is controlled or configured to perform the processing method of the flowchart described in the method embodiment of a fire detection method of this application, such as the terminal and computer program product of the above flowchart, falls within the scope of the related products described in this application.

[0187] Obviously, those skilled in the art can make various modifications and variations to the fire detection method, apparatus, electronic device, and storage medium provided in this application without departing from the spirit and scope of this application. Therefore, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application also intends to include these modifications and variations.

Claims

1. A fire detection method, characterized in that, The method is applied to an edge computing gateway in a fire detection system, which also includes a detection terminal. The method includes: Construction information is obtained through an interface connection with the construction management platform; An interference factor is generated based on the construction information, and the interference factor is used to characterize the degree of interference to fire detection. The sensor data used for fire detection is obtained from the detection terminal; The fire detection result is determined based on the sensor data and the interference factor. The construction information includes a list of work plans for the current floor and adjacent floors above and below. Each work plan list includes one or more work plans, and each work plan includes a work type, three-dimensional coordinates, a start time, and an end time. When the work plan list includes only one work plan, the generation of interference factors based on the construction information includes: The weight of a task type is determined based on a preset mapping table and the task type. The preset mapping table includes the correspondence between task types and task type weights. The higher the probability that a task type will cause a false fire alarm, the higher the weight of that task type. Based on the distance between the three-dimensional coordinate position and the position of the detection terminal, a spatial distance weight is determined; the smaller the distance, the greater the spatial distance weight. Based on the inclusion relationship between the first time interval and the first time, the weight of the time window is determined. The first time interval is determined based on the start time of the plan and the end time of the plan. The first time is the time when the construction information is obtained. The interference factor is generated based on the job type weight, the spatial distance weight, and the time window weight.

2. The method according to claim 1, characterized in that, When the work plan list includes multiple work plans, the generation of interference factors based on the construction information includes: Obtain a first interference factor for each of the plurality of work plans, wherein the first interference factor is determined based on the work type, the three-dimensional coordinate position, the plan start time, and the plan end time included in each work plan; The interference factor is generated based on multiple first interference factors.

3. The method according to claim 2, characterized in that, The generation of the interference factor based on multiple first interference factors includes: The first operation value is obtained by performing a cube operation on each of the plurality of first interference factors and then summing the results. The summation of the plurality of first interference factors yields the second calculated value; The interference factor is obtained by taking the square root of the ratio between the first calculated value and the second calculated value.

4. The method according to any one of claims 1-3, characterized in that, The process of determining the fire detection result based on the sensor data and the interference factor includes: A feature vector is constructed based on the sensor data and the interference factor; The feature vector is input into the classification model to obtain the probabilities corresponding to the first classification result, the second classification result, and the third classification result, respectively. The first classification result is used to characterize no fire, the second classification result is used to characterize a suspected fire, and the third classification result is used to characterize a confirmed fire. The fire detection result is determined based on the probabilities corresponding to the first classification result, the second classification result, and the third classification result, respectively.

5. The method according to claim 4, characterized in that, The process of determining the fire detection result based on the probabilities corresponding to the first classification result, the second classification result, and the third classification result includes: If the probability of the third classification result is greater than or equal to the first threshold, then the fire detection result is determined to be a confirmed fire. If the probability of the third classification result is less than the first threshold, and the probability of the second classification result is greater than or equal to the second threshold, then the fire detection result is determined to be a suspected fire. If the probability of the third classification result is less than the first threshold, the probability of the second classification result is less than the second threshold, and the probability of the first classification result is greater than or equal to the third threshold, then the fire detection result is determined to be no fire. If the probability of the third classification result is less than the first threshold, the probability of the second classification result is less than the second threshold, and the probability of the first classification result is less than the third threshold, then the fire detection result is determined to be a suspected fire.

6. The method according to claim 5, characterized in that, The fire detection system also includes an alarm, and the method further includes: If the fire detection result confirms a fire, the alarm will be activated to sound a fire alarm.

7. The method according to claim 5, characterized in that, The edge computing gateway is connected to a cloud server, and the method further includes: If the fire detection result is a suspected fire, a review request is sent to the cloud server, which is used to instruct the review of the fire detection result.

8. A fire detection device, characterized in that, The device is used in an edge computing gateway of a fire detection system, which also includes a detection terminal. The device comprises: The communication unit is used to obtain construction information through an interface connection with the construction management platform; A processing unit is used to generate an interference factor based on the construction information, the interference factor being used to characterize the degree of interference to fire detection; The communication unit is also used to acquire sensor data for fire detection from the detection terminal; The processing unit is also used to determine the fire detection result based on the sensor data and the interference factor; The construction information includes a list of work plans for the current floor and adjacent floors above and below. Each work plan list includes one or more work plans, and each work plan includes a work type, three-dimensional coordinate position, planned start time, and planned end time. When the work plan list includes only one work plan, in the aspect of generating interference factors based on the construction information, the processing unit is specifically used for: The weight of a task type is determined based on a preset mapping table and the task type. The preset mapping table includes the correspondence between task types and task type weights. The higher the probability that a task type will cause a false fire alarm, the higher the weight of that task type. Based on the distance between the three-dimensional coordinate position and the position of the detection terminal, a spatial distance weight is determined; the smaller the distance, the greater the spatial distance weight. Based on the inclusion relationship between the first time interval and the first time, the weight of the time window is determined. The first time interval is determined based on the start time of the plan and the end time of the plan. The first time is the time when the construction information is obtained. The interference factor is generated based on the job type weight, the spatial distance weight, and the time window weight.

9. An electronic device, the device comprising a processor, a memory, and executable program code stored in the memory, characterized in that, The processor is configured to retrieve the executable program code stored in the memory to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-7.

Images