Rapid picking, monitoring and early warning system for incentive parameters of non-coal mine exogenous fire

Abstract

The invention provides a non-coal mine exogenous fire hazard incentive parameter rapid picking and monitoring early warning system, which realizes panoramic picking of non-coal mine exogenous fire hazard core incentive parameters through combination of a network camera, a thermal imaging special series cylinder machine, two types of four-in-one gas sensors and an early warning software platform. A localized processing architecture with an edge computing box as a core is adopted, data acquisition, analysis and decision links are completed at the front end in a concentrated mode, remote computing of a ground central station is not needed, data transmission delay is avoided, response time is shortened, the edge computing box fuses video, infrared and gas multi-source data through an algorithm, the recognition accuracy rate is increased to 95% or above, and the recognition efficiency is improved. According to the scheme, the speed and precision of the fire detection method are improved for the areas with high fire risk, such as vehicles, cables and power transformation and distribution chambers in the transportation roadway in the non-coal underground mine, and the problem that a traditional fire monitoring system is poor in real-time performance is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fire monitoring, in particular to a non-coal mine external cause fire inducement parameter rapid picking and monitoring and early warning system. BACKGROUND

[0002] Non-coal mine external cause fire is mainly caused by electrical faults, mechanical friction, blasting operations, welding sparks, etc. For example, electrical equipment overload or cable aging may cause short circuit fire, and the risk of high temperature igniting sulfide dust in blasting operation is significant.

[0003] At present, the detection of non-coal mine external cause fire mainly uses a single gas sensor or a thermal imaging special series cylinder machine to identify and alarm the gas or temperature parameters in the non-coal mine underground. There is a single monitoring parameter, which can only monitor and warn a single parameter. The present application fuses temperature and eight gas parameters, proposes to identify underground fire based on multi-source information fusion, uses a normalized weighting algorithm to fuse video detection information and multi-sensor detection information, and gives a dynamic adjustment strategy of evaluation index. SUMMARY

[0004] The present application provides a non-coal mine external cause fire inducement parameter rapid picking and monitoring and early warning system to solve the problems proposed in the background art.

[0005] A non-coal mine external cause fire inducement parameter rapid picking and monitoring and early warning system, comprising: an edge computing box, a plurality of network cameras, a plurality of thermal imaging special series cylinder machines and two types of four-in-one gas sensors; The network cameras, thermal imaging special series cylinder machines and two types of four-in-one gas sensors are connected to the edge computing box; The edge computing box is used for algorithm analysis based on the collected data of the network cameras, thermal imaging special series cylinder machines and two four-in-one gas sensors, determines fire identification data, and generates fire warning information; Further comprising: a system platform for remotely receiving fire warning information, generating alarm information based on the fire warning information, and transmitting the alarm information to the sensing instrument for alarm.

[0006] Preferably, the two four-in-one gas sensors are respectively a four-in-one gas sensor composed of CO, SO2, NH3 and CO2 four gases and a four-in-one gas sensor composed of O2, CH4, NO2 and H2S four gases.

[0007] Preferably, the network camera is used for collecting video data of the non-coal mine monitoring area; The thermal imaging special series cylinder machine is used for collecting infrared imaging data of a non-coal mine monitoring area; and the four-in-one gas sensor is used for collecting concentration data of CO, SO2, NH3, CO2, O2, CH4, NO2 and H2S in the non-coal mine monitoring area.

[0008] Preferably, the edge computing box comprises: a data processing unit configured to identify and process the collected data to obtain target collection data; a fire identification unit configured to detect a flame based on the target collection data in combination with an identification algorithm to obtain a fire area and a confidence level; a fire warning unit configured to determine fire warning information based on gas concentration information and temperature information of the fire area in combination with the confidence level.

[0009] Preferably, the data processing unit comprises: a video data processing unit configured to process video data based on a filtering algorithm and extract a region of interest from the processed video data based on a pre-standard monitoring area to obtain target video data; an infrared imaging data processing unit configured to mark infrared imaging pixel points higher than a preset temperature threshold as high-temperature points, and remove isolated points smaller than a preset area from the high-temperature points to obtain target infrared imaging data; a concentration data processing unit configured to remove outliers from the concentration data and filter concentration values greater than each gas safety threshold as target concentration data.

[0010] Preferably, the fire identification unit comprises: a video analysis unit configured to select data with pixels within a preset range from the target video data as initial flame features, identify the initial flame features based on a flame identification model to obtain flame area and flame color information; an infrared imaging analysis unit configured to determine a temperature gradient calculation based on the target infrared imaging data, and determine a temperature change rate based on the temperature gradient calculation result; a concentration analysis unit configured to perform association analysis on the target concentration data based on a preset gas combination association rule to output fire gas features; an alignment unit configured to align the flame area and flame color information, the temperature change rate and the fire gas features in time stamp and space region, and select fire data information pointing to the same time or space region according to the alignment result; a region determination unit configured to determine a fire area based on the fire data information. The confidence determination unit is used to configure video weights, infrared imaging weights, and concentration weights for fire data information, and calculate the confidence level of the fire area by combining the matching degree of time or spatial region. Preferably, the confidence level determination unit includes: The weight allocation unit is used to set video weight, infrared imaging weight and concentration weight based on dust interference from video data, imaging accuracy of infrared imaging data and combined correlation of concentration data. The correction determination unit is used to set a first correction coefficient when the fire data information does not match in time but matches in space, set a second correction coefficient when the fire data information matches in time but does not match in space, and set a third correction coefficient when the fire data information matches in time and matches in space. The calculation unit is used to obtain an initial confidence level based on the individual confidence level of fire data information, combined with video weight, infrared imaging weight and concentration weight. The confidence level of the fire area is obtained based on the product of the initial confidence level and the correction coefficient.

[0011] Preferably, the fire early warning unit includes: The rule-establishing unit is used to establish an early warning matching table based on setting four warning levels: red, orange, yellow, and blue, and configuring corresponding concentration ranges, temperature ranges, and confidence ranges for each warning level; The early warning generation unit is used to match the gas concentration and temperature information of the fire area with the early warning matching table based on the confidence level, and generate fire early warning information based on the early warning level and the fire area.

[0012] Preferably, the system platform is also used to display the collected data and fire early warning information.

[0013] Compared with the prior art, the present invention has achieved the following beneficial effects: By combining network cameras, special series thermal imaging cylinder cameras, and two types of four-in-one gas sensors, a panoramic acquisition of core parameters of externally induced fires in non-coal mines is achieved. The localized processing architecture, centered on an edge computing box, centralizes data acquisition, analysis, and decision-making at the front end, eliminating reliance on remote computing at a ground-based central station, avoiding data transmission delays, and improving response time. The edge computing box integrates video, infrared, and gas multi-source data through algorithms, increasing the identification accuracy to over 95% and reducing false alarm rates. This solution targets high-risk fire areas in non-coal underground mines, such as transportation roadways, vehicles, cables, and power distribution chambers, improving the speed and accuracy of fire detection methods and addressing the poor real-time performance of traditional fire monitoring systems. The system platform allows monitoring personnel to intuitively view real-time monitoring data and fire early warning information for more accurate analysis and management of the underground environment.

[0014] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0015] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a structural diagram of a system for rapid acquisition, monitoring and early warning of external fire causation parameters in non-coal mines, as described in an embodiment of the present invention. Figure 2 This is a structural diagram of the edge computing box described in an embodiment of the present invention. Detailed Implementation

[0017] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0018] Example 1: This invention provides a system for rapid acquisition, monitoring, and early warning of external fire causative parameters in non-coal mines, such as... Figure 1 As shown, it consists of: an edge computing box, several network cameras, several special series thermal imaging tubes, and two types of four-in-one gas sensors; The network camera, the special series thermal imaging tube camera, and the two types of four-in-one gas sensors are respectively connected to the edge computing box; The edge computing box is used to perform algorithm analysis on data collected from network cameras, special series thermal imaging cylinders, and two four-in-one gas sensors to determine fire identification data and generate fire early warning information. It also includes: a system platform for remotely receiving fire early warning information, generating alarm information based on the fire early warning information, and transmitting the alarm information to sensing instruments for alarm activation.

[0019] In this embodiment, such as Figure 1 As shown, it also includes a power supply, a host computer for real-time monitoring and early warning, and an alarm built into the four-in-one gas sensor.

[0020] In this embodiment, the edge computing box is an intelligent recognition board (hereinafter referred to as the AI ​​board) based on the domestically produced Cambricon MLU220 intelligent chip. The overall project is divided into two parts: one is a general-purpose intrinsically safe mining edge processing board, and the other is an application for underground fire source identification based on the edge processing board. The two projects share the AI ​​board hardware. The AI ​​board has the ability to recognize targets and is compatible with other intelligent algorithms, meeting the needs of operator support and computational efficiency for mixed tasks. It supports real-time image data acquisition, target recognition, and high-speed communication and data interaction with external devices. It also provides standardized software and hardware calling services for top-level AI algorithms, and provides technical support for different development stages such as software development and debugging, simulation verification, and specific application scenarios.

[0021] In this embodiment, the fixed four-in-one gas sensor can be used in metallurgy, coking, petrochemical, environmental monitoring, and other fields to continuously monitor the concentration of toxic, harmful, and explosive gases in the environment. An alarm can be triggered when the gas concentration exceeds the limit.

[0022] In this embodiment, the webcam can be used in locations with explosive mixtures of flammable gases, vapors and air, or in locations with explosive hazards formed by mixtures of flammable dust and air.

[0023] In this embodiment, the special series of thermal imaging bullet cameras supports intelligent functions such as temperature measurement, intelligent analysis, smoke and fire detection or dynamic fire point search, and AI platform. It also supports fog penetration, strong light suppression, Smart IR anti-infrared overexposure technology, low bit rate, ROI region of interest enhancement coding, SVC adaptive coding technology, and supports events such as motion detection, occlusion alarm, and audio anomaly detection.

[0024] The beneficial effects of the above design scheme are as follows: By combining network cameras, special series thermal imaging cylinder cameras, and two types of four-in-one gas sensors, a panoramic acquisition of core inducing parameters of external fires in non-coal mines is achieved. The localized processing architecture with edge computing boxes as the core centralizes data acquisition, analysis, and decision-making at the front end, eliminating the need for remote computing at ground central stations, avoiding data transmission delays, and improving response time. The edge computing boxes integrate video, infrared, and gas multi-source data through algorithms, increasing the recognition accuracy to over 95% and reducing the false alarm rate. This scheme is designed for areas with high fire risk in non-coal underground mines, such as vehicles, cables, and power distribution chambers in transport roadways, improving the speed and accuracy of fire detection methods and solving the problem of poor real-time performance in traditional fire monitoring systems. Through the system platform, monitoring personnel can intuitively see real-time monitoring data and fire early warning information, enabling more accurate analysis and management of the underground environment.

[0025] Example 2: Based on Example 1, this embodiment of the invention provides a rapid acquisition and monitoring early warning system for external fire causation parameters in non-coal mines, comprising: two four-in-one gas sensors, one consisting of CO, SO2, NH3, and CO2, and the other consisting of O2, CH4, NO2, and H2S.

[0026] The beneficial effects of the above design scheme are: the gas sensor configuration scheme provides highly reliable and targeted technical support for early gas warning of external fires in non-coal mines through four major advantages: accurate coverage of characteristic gases, avoidance of detection interference, multi-parameter collaborative verification, and compliance adaptation. It directly serves the rapid judgment of fire causes and emergency response decisions.

[0027] Example 3: Based on Example 1, this embodiment of the invention provides a rapid acquisition, monitoring, and early warning system for external fire causation parameters in non-coal mines, comprising: The network camera is used to collect video data from non-coal mine monitoring areas; The special series of thermal imaging tubes is used to collect infrared imaging data of non-coal mine monitoring areas; the four-in-one gas sensor is used to collect concentration data of CO, SO2, NH3, CO2 and O2, CH4, NO2 and H2S in non-coal mine monitoring areas.

[0028] In this embodiment, the concentration data of CO, SO2, NH3, CO2 and O2, CH4, NO2 and H2S from video data and infrared imaging data are uploaded to the edge computing box as collected data.

[0029] The beneficial effects of the above design scheme are: the edge computing box integrates video, infrared and gas multi-source data through algorithms, which improves the recognition accuracy to more than 95% and reduces the false alarm rate.

[0030] Example 4: Based on Example 1, this embodiment of the invention provides a rapid acquisition, monitoring, and early warning system for external fire causation parameters in non-coal mines, such as... Figure 2 As shown, the edge computing box includes: The data processing unit is used to identify and process the collected data to obtain the target collected data; The fire identification unit is used to detect flames based on target acquisition data and identification algorithms to obtain the fire area and confidence level. The fire early warning unit is used to determine fire early warning information based on gas concentration and temperature information in the fire area, combined with confidence level.

[0031] The beneficial effects of the above design scheme are as follows: by processing the raw collected data, the data quality is guaranteed, reducing the risk of misjudgment in subsequent identification and early warning from the source; based on the flame detection and confidence calculation of the target collected data, multimodal fusion identification breaks through the limitations of single data identification and improves the accuracy of fire identification; by weighted fusion of single-modal features of video, infrared and gas, confidence is output to provide a quantitative basis for early warning classification; by generating early warning information based on gas concentration, temperature and confidence, the problems of single monitoring and early warning threshold and rigid response methods in traditional monitoring and early warning are solved, providing reliable data information for the early prevention and control of external fires in mines.

[0032] Example 5: Based on Embodiment 4, this embodiment of the invention provides a rapid acquisition, monitoring, and early warning system for external fire causation parameters in non-coal mines. The data processing unit includes: The video data processing unit is used to process video data based on filtering algorithms and extract the region of interest from the processed video data based on a pre-defined standard monitoring area to obtain the target video data. An infrared imaging data processing unit is used to select infrared imaging pixels with a temperature higher than a preset temperature threshold and mark them as high-temperature points, and remove isolated points with an area smaller than a preset area from the high-temperature points to obtain target infrared imaging data. The concentration data processing unit is used to remove outliers from the concentration data and select concentration values ​​that are greater than the safety threshold of each gas as target concentration data.

[0033] In this embodiment, the filtering algorithm is one or both of Gaussian filtering and median filtering.

[0034] In this embodiment, target video data, target infrared imaging data, and target infrared imaging data together constitute target acquisition data.

[0035] The beneficial effects of the above design scheme are as follows: By specifically processing three types of data—video, infrared, and gas concentration—this design scheme achieves the preprocessing goals of noise filtering, feature focusing, and data simplification. Its core benefits lie in adapting to the complex environment of non-coal mines, improving data quality, and providing accurate input for subsequent fire identification. The video data processing unit specifically eliminates mine environmental noise and focuses on high-risk areas; the infrared imaging data processing unit accurately filters fire-related high temperatures and eliminates interfering heat sources; and the concentration data processing unit purifies gas data and focuses on dangerous signals. This provides crucial support for the accurate identification and rapid early warning of the entire monitoring and early warning system.

[0036] Example 6: Based on Embodiment 5, this embodiment of the invention provides a rapid acquisition, monitoring, and early warning system for external fire causation parameters in non-coal mines. The fire identification unit includes: The video analysis unit is used to select data of pixels within a preset range from the target video data as initial flame features, and to identify the initial flame features based on the flame recognition model to obtain flame area and flame color information. The infrared imaging analysis unit is used to determine the temperature gradient calculation based on the target infrared imaging data, and to determine the temperature change rate based on the temperature gradient calculation result. The concentration analysis unit is used to perform correlation analysis on target concentration data based on preset gas combination correlation rules, and output the fire gas characteristics. The alignment unit is used to perform timestamp alignment and spatial region alignment of flame area and flame color information, temperature change rate and fire gas characteristics, and select fire data information pointing to the same time or spatial region based on the alignment results. The area determination unit is used to determine the fire area based on fire data information; The confidence determination unit is used to configure video weights, infrared imaging weights, and concentration weights for fire data information, and calculate the confidence level of the fire area by combining the matching degree of time or spatial region.

[0037] The beneficial effects of the above design scheme are: accurate extraction of multimodal features from video, infrared, and gas, covering all dimensions of fire characteristics; time stamp alignment and spatial region alignment, ensuring that multimodal data points to fire events in the same time and space, solving the problem of misassociation caused by data asynchrony and regional mismatch in non-coal mine monitoring; and based on the determination of fire area and confidence level, realizing the upgrade of fire identification from qualitative judgment to quantitative description, providing a quantitative basis for the graded response of subsequent early warning units.

[0038] Example 7: Based on Embodiment 6, this embodiment of the invention provides a rapid acquisition, monitoring, and early warning system for external fire causation parameters in non-coal mines. The confidence level determination unit includes: The weight allocation unit is used to set video weight, infrared imaging weight and concentration weight based on dust interference from video data, imaging accuracy of infrared imaging data and combined correlation of concentration data. The correction determination unit is used to set a first correction coefficient when the fire data information does not match in time but matches in space, set a second correction coefficient when the fire data information matches in time but does not match in space, and set a third correction coefficient when the fire data information matches in time and matches in space. The calculation unit is used to obtain an initial confidence level based on the individual confidence level of fire data information, combined with video weight, infrared imaging weight and concentration weight. The confidence level of the fire area is obtained based on the product of the initial confidence level and the correction coefficient.

[0039] In this embodiment, the sum of the video weight, infrared imaging weight, and concentration weight is 1. The greater the dust interference, the smaller the video weight and the higher the imaging accuracy. The greater the infrared imaging weight, the greater the correlation between the combined concentration data and the greater the concentration weight.

[0040] In this embodiment, the first correction factor is set to 0.8, the second correction factor is set to 0.6, and the third correction factor is set to 1.

[0041] In this embodiment, the individual confidence level of fire data information is determined based on the relationship between information features and a pre-matching table.

[0042] In this embodiment, the correction factor is one of the first correction factor, the second correction factor, and the third correction factor.

[0043] The beneficial effects of the above design scheme are: by combining the reliability differences of the three types of data in the mining environment to set weights, the weights are made more in line with the actual scenario. In response to the possible time asynchrony and spatial mismatch problems of multi-source data in non-coal mines, the spatiotemporal consistency is quantified through three-level correction coefficients, which significantly improves the authenticity of confidence.

[0044] Example 8: Based on Embodiment 7, this embodiment of the invention provides a rapid acquisition, monitoring, and early warning system for external fire causation parameters in non-coal mines. The fire early warning unit includes: The rule-establishing unit is used to establish an early warning matching table based on setting four warning levels: red, orange, yellow, and blue, and configuring corresponding concentration ranges, temperature ranges, and confidence ranges for each warning level; The early warning generation unit is used to match the gas concentration and temperature information of the fire area with the early warning matching table based on the confidence level, and generate fire early warning information based on the early warning level and the fire area.

[0045] In this embodiment, the warning levels of red, orange, yellow, and blue decrease sequentially.

[0046] The beneficial effects of the above design scheme are as follows: through the design of four-level graded early warning, quantitative matching rules, and precise regional correlation, the fire identification results are transformed into early warning information that can directly guide mine emergency response. Its core beneficial effects are to achieve precise risk classification, standardized response decision-making, and efficient emergency response, and to specifically solve the problems of no classification, difficulty in implementation, and chaotic response in non-coal mine fire early warning.

[0047] Example 9: Based on Embodiment 1, this embodiment of the invention provides a rapid acquisition and monitoring early warning system for external fire causation parameters in non-coal mines. The system platform is characterized in that it is also used to display the collected data and fire early warning information.

[0048] In this embodiment, the left-hand interface displays sensor data processed by the edge computing box, showing real-time concentration information of eight gases monitored in the mine. These gas concentrations are divided into different levels so that monitoring personnel can intuitively understand the changes in underground gas concentrations and take corresponding prevention and control measures based on the early warning results. This real-time monitoring and automatic early warning function ensures the accuracy and timeliness of the data, providing strong support for safety management. The upper right-hand interface displays line graphs showing the changes in the concentrations of these eight gases over time, allowing monitoring personnel to intuitively see the current and historical fluctuations in gas concentrations. Users can also select specific gas concentrations to query for more precise analysis and management of the underground environment. The lower right-hand interface displays video content from infrared cameras and network cameras processed by the edge computing box. The edge computing box identifies and processes the video, using algorithms to identify flames detected in the video and indicate their confidence level.

[0049] The beneficial effects of the above design scheme are that it enables monitoring personnel to intuitively see real-time monitoring data and fire early warning information, so as to more accurately analyze and manage the underground environment.

[0050] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this application and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A rapid acquisition, monitoring, and early warning system for external fire-causing parameters in non-coal mines, characterized in that, include: It consists of an edge computing box, several network cameras, several special series thermal imaging tubes, and two types of four-in-one gas sensors; The network camera, the special series thermal imaging tube camera, and the two types of four-in-one gas sensors are respectively connected to the edge computing box; The edge computing box is used to perform algorithm analysis on data collected from network cameras, special series thermal imaging cylinders, and two four-in-one gas sensors to determine fire identification data and generate fire early warning information. It also includes a system platform for remotely receiving fire early warning information, generating alarm information based on the fire early warning information, and transmitting the alarm information to sensing instruments for alarm activation.

2. The rapid acquisition, monitoring, and early warning system for external fire causation parameters in non-coal mines according to claim 1, characterized in that, include: The two four-in-one gas sensors are a four-in-one gas sensor composed of four gases: CO, SO2, NH3, and CO2, and a four-in-one gas sensor composed of four gases: O2, CH4, NO2, and H2S.

3. The rapid acquisition, monitoring, and early warning system for external fire causation parameters in non-coal mines according to claim 1, characterized in that: The network camera is used to collect video data from non-coal mine monitoring areas; The special series of thermal imaging tubes is used to collect infrared imaging data of non-coal mine monitoring areas; the four-in-one gas sensor is used to collect concentration data of CO, SO2, NH3, CO2 and O2, CH4, NO2 and H2S in non-coal mine monitoring areas.

4. The rapid acquisition, monitoring, and early warning system for external fire causation parameters in non-coal mines according to claim 1, characterized in that, The edge computing box includes: The data processing unit is used to identify and process the collected data to obtain the target collected data; The fire identification unit is used to detect fire points based on target acquisition data and identification algorithms, and to obtain the fire area and confidence level. The fire early warning unit is used to determine fire early warning information based on gas concentration and temperature information in the fire area, combined with confidence level.

5. The rapid acquisition, monitoring, and early warning system for external fire causation parameters in non-coal mines according to claim 4, characterized in that, The data processing unit includes: The video data processing unit is used to process video data based on filtering algorithms and extract the region of interest from the processed video data based on a pre-defined standard monitoring area to obtain the target video data. An infrared imaging data processing unit is used to select infrared imaging pixels with a temperature higher than a preset temperature threshold and mark them as high-temperature points, and remove isolated points with an area smaller than a preset area from the high-temperature points to obtain target infrared imaging data. The concentration data processing unit is used to remove outliers from the concentration data and select concentration values ​​that are greater than the safety threshold of each gas as target concentration data.

6. The rapid acquisition, monitoring, and early warning system for external fire causation parameters in non-coal mines according to claim 5, characterized in that, The fire detection unit includes: The video analysis unit is used to select data of pixels within a preset range from the target video data as initial flame features, and to identify the initial flame features based on the flame recognition model to obtain flame area and flame color information. The infrared imaging analysis unit is used to determine the temperature gradient calculation based on the target infrared imaging data, and to determine the temperature change rate based on the temperature gradient calculation result. The concentration analysis unit is used to perform correlation analysis on target concentration data based on preset gas combination correlation rules, and output the fire gas characteristics. The alignment unit is used to perform timestamp alignment and spatial region alignment of flame area and flame color information, temperature change rate and fire gas characteristics, and select fire data information pointing to the same time or spatial region based on the alignment results. The area determination unit is used to determine the fire area based on fire data information; The confidence determination unit is used to configure video weights, infrared imaging weights, and concentration weights for fire data information, and calculate the confidence level of the fire area by combining the matching degree of time or spatial region.

7. The rapid acquisition, monitoring, and early warning system for external fire causation parameters in non-coal mines according to claim 6, characterized in that, The confidence level determination unit includes: The weight allocation unit is used to set video weight, infrared imaging weight and concentration weight based on dust interference from video data, imaging accuracy of infrared imaging data and combined correlation of concentration data. The correction determination unit is used to set a first correction coefficient when the fire data information does not match in time but matches in space, set a second correction coefficient when the fire data information matches in time but does not match in space, and set a third correction coefficient when the fire data information matches in time and matches in space. The calculation unit is used to obtain an initial confidence level based on the individual confidence level of fire data information, combined with video weight, infrared imaging weight and concentration weight. The confidence level of the fire area is obtained based on the product of the initial confidence level and the correction coefficient.

8. A rapid acquisition, monitoring, and early warning system for external fire causation parameters in non-coal mines according to claim 7, characterized in that, The fire early warning unit includes: The rule-establishing unit is used to establish an early warning matching table based on setting four warning levels: red, orange, yellow, and blue, and configuring corresponding concentration ranges, temperature ranges, and confidence ranges for each warning level; The early warning generation unit is used to match the gas concentration and temperature information of the fire area with the early warning matching table based on the confidence level, and generate fire early warning information based on the early warning level and the fire area.

9. A rapid acquisition, monitoring, and early warning system for external fire causation parameters in non-coal mines according to claim 1, characterized in that, The system platform is also used to display collected data and fire early warning information.

Images