Fire disaster intelligent monitoring and early warning method and system based on infrared thermal imaging technology

Abstract

The invention relates to the technical field of image data processing, in particular to a fire disaster intelligent monitoring and early warning method and system based on an infrared thermal imaging technology, and the method comprises the steps: obtaining continuous multi-frame thermal infrared images in a set time window of a to-be-monitored region; processing the continuous multi-frame thermal infrared image by using an adaptive threshold segmentation algorithm to obtain a plurality of hot spot regions, and obtaining a target hot spot region of the target hot spot region in the continuous multi-frame thermal infrared image by using a target detection algorithm; the target hot spot area is any hot spot area in continuous multi-frame thermal infrared images; and calculating the fire confidence coefficient of the target hot spot area, and triggering early warning when the fire confidence coefficient exceeds a preset threshold value. According to the invention, the problem of low accuracy of fire risk monitoring and early warning in practical application is solved.

Description

Technical Field

[0001] This invention relates to the field of image data processing technology. More specifically, this invention relates to a method and system for intelligent monitoring and early warning of fire disasters based on infrared thermal imaging technology. Background Technology

[0002] With the acceleration of industrialization, the expansion of urban construction, and the continuous increase in energy utilization density, fire, as a disaster with strong suddenness, great destructive power, and wide-ranging social impact, has become an important issue that urgently needs to be addressed in various public safety systems and industrial safety management.

[0003] In recent years, with the development of infrared thermal imaging technology, non-contact, real-time, and long-distance monitoring of object surface temperature has gradually become an important research direction in the field of fire early warning. Existing fire monitoring methods based on infrared thermal imaging mainly acquire single-frame thermal infrared images or short-time image sequences, identify high-temperature pixels or local hotspot areas in the images, and combine temperature thresholds, pixel area changes, or simple hotspot growth patterns to determine the fire situation. For example, some methods use a fixed temperature threshold to detect areas in the image with temperatures higher than the background temperature, considering pixels with temperatures exceeding the threshold as potential fire sources, thus triggering an alarm; other methods calculate the rate of change of hotspot area or the increasing trend of the average temperature in consecutive frames to determine whether the heat source is persistent or expanding. These methods, to a certain extent, improve the monitoring capability of large areas, enabling the system to detect some abnormal heat sources in the early stages of a fire, thus possessing higher early warning sensitivity and coverage than traditional sensor methods.

[0004] However, existing infrared thermal imaging-based monitoring methods still have significant shortcomings. First, most methods only focus on the temperature distribution of a single frame image or a local time series, lacking a comprehensive assessment of the spatial location changes of hotspots in consecutive frames, the persistence of heat, and the trend of regional expansion. This results in insufficient ability to distinguish between real fire sources and environmental interference. Second, existing methods typically lack persistent analysis of weak heat sources in the early stages of a fire, relying solely on instantaneous pixel temperature values ​​or area changes. This fails to fully reflect the physical evolution characteristics of the fire's latent stage and has limited ability to identify smoldering fires or the early stages of slow fire source expansion, leading to low accuracy in fire risk monitoring and early warning in practical applications. Summary of the Invention

[0005] To address the problem of low accuracy in fire risk monitoring and early warning in practical applications as mentioned in the background section, this invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides a method for intelligent monitoring and early warning of fire disasters based on infrared thermal imaging technology, comprising: acquiring multiple consecutive frames of thermal infrared images of a monitored area within a set time window; processing the multiple consecutive frames of thermal infrared images using an adaptive threshold segmentation algorithm to obtain multiple hotspot areas, and acquiring the target hotspot area in the multiple consecutive frames of thermal infrared images using a target detection algorithm; the target hotspot area being any hotspot area in the multiple consecutive frames of thermal infrared images; calculating the fire confidence level of the target hotspot area, and triggering an early warning when the fire confidence level exceeds a preset threshold; the fire confidence level is positively correlated with the expansion degree of the target hotspot area and negatively correlated with the fluctuation degree of the target hotspot area; the expansion degree is positively correlated with the thermal persistence of the target hotspot area and the pixel change of the target hotspot area within the set time window; the thermal persistence is positively correlated with the pixel temperature of the target hotspot area in the last frame of thermal infrared images within the set time window and negatively correlated with the pixel temperature of the target hotspot area in the first frame of thermal infrared images within the set time window and the linear regression slope of the pixel temperature of the target hotspot area in the multiple consecutive frames of thermal infrared images within the set time window as a function of time.

[0007] The above technical solution can give high weight to abnormal heat sources that are continuously growing and located stably, while automatically reducing the weight of hot spots that fluctuate greatly or appear only briefly. This significantly improves the accuracy, real-time performance and robustness of early fire warnings, providing reliable data support and decision-making basis for fire monitoring in complex environments.

[0008] Furthermore, the first Fire confidence in key areas for: , For the first The extent of expansion in each hotspot area Using natural constants An exponential function with base 0. For the first The degree of fluctuation in each hotspot area This is the threshold for the degree of fluctuation.

[0009] The aforementioned technical solution combines the spatial expansion characteristics and location stability of hotspot areas to comprehensively assess fire risk, enabling fire confidence to simultaneously reflect whether the heat source is continuously expanding and whether it remains relatively stable. When the hotspot shows significant expansion and location stability, the confidence value increases significantly, thereby enhancing the ability to identify real, continuously heating fire sources. Conversely, when the hotspot exhibits significant spatial fluctuations or random variations, even with a certain expansion trend, the exponential decay mechanism automatically reduces the confidence value, effectively suppressing false alarms caused by environmental interference, sensor noise, or short-term hotspots.

[0010] Furthermore, the first Expansion of hotspot areas for: , For the first The duration of heat in each hotspot area To find the maximum value function, , The first The hotspot area in the first Frame, First The total number of pixels in a frame of thermal infrared image. To preset hyperparameters, This is used to set the total number of consecutive thermal infrared images within a time window.

[0011] The above technical solution combines the thermal persistence of the hot spot area with its area expansion characteristics in a continuous thermal infrared image sequence, which can highlight the real, continuously heating and expanding fire source, while suppressing the interference of area reduction or random changes caused by environmental disturbances, sensor noise or local pixel fluctuations on the results.

[0012] Furthermore, the first The duration of heat in each hotspot area for: , , These are the last frame and the first frame of the thermal infrared image within the set time window, respectively. Pixel temperature in hotspot areas , The first Frame, First The first frame of the thermal infrared image Pixel temperature in hotspot areas Using natural constants An exponential function with base 0. To set the total number of consecutive thermal infrared images within a time window, To define the first frame of multiple consecutive thermal infrared images within a set time window The slope of a linear regression of pixel temperature changes over time in hotspot regions. This is the preset first hyperparameter.

[0013] The above technical solution, through comprehensive analysis of the temperature change amplitude, cumulative fluctuation, and overall trend of hotspot areas in continuous thermal infrared image sequences, can distinguish between real heat sources with stable temperature increases or long-term heat release and pseudo-hotspots with short-term fluctuations and random rises and falls. At the same time, by adjusting the temperature change trend, the indicators of continuously heating hotspots are strengthened, while unstable or decaying hotspots are suppressed. This effectively improves the accuracy of early identification of abnormal heat sources and the robustness of the system to environmental interference, providing a solid data foundation for reliable early warning of fires or abnormal hotspots.

[0014] Furthermore, the first The degree of fluctuation in each hotspot area for: , , For the first The hotspot area in the first Frame, First Centroid coordinates in a frame of thermal infrared image To set the total number of consecutive thermal infrared images within a time window, It is the Euclidean norm.

[0015] The aforementioned technical solution effectively suppresses the impact of instantaneous shifts caused by sensor noise, environmental disturbances, or light reflection on monitoring results, and can accurately distinguish between real, persistent heat sources and short-term, randomly moving pseudo-hot spots. Furthermore, through quantitative assessment of hot spot location fluctuations, this solution provides a reliable basis for subsequent hot spot expansion analysis and fire confidence calculation, improving the accuracy of early fire anomaly identification and the robustness of monitoring.

[0016] Furthermore, an infrared thermal imager is used to acquire multiple consecutive frames of thermal infrared images of the area to be monitored within a set time window.

[0017] Furthermore, the specific steps for obtaining the target hotspot region in consecutive frames of thermal infrared images using the target detection algorithm are as follows: calculate the IoU value between the target hotspot region and multiple hotspot regions in consecutive frames of thermal infrared images, and take the hotspot region with the largest IoU value as the target hotspot region in consecutive frames of thermal infrared images.

[0018] Furthermore, the pixel temperature of the target hotspot area is the sum of the pixel temperatures within the target hotspot area.

[0019] Furthermore, the continuous multi-frame thermal infrared images are subjected to noise reduction and contrast enhancement processing.

[0020] In a second aspect, the present invention provides a fire disaster intelligent monitoring and early warning system based on infrared thermal imaging technology, including a memory and a processor. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the fire disaster intelligent monitoring and early warning method based on infrared thermal imaging technology described above is implemented.

[0021] The beneficial effects of this invention are as follows: This invention combines infrared thermal imaging technology with continuous frame image analysis, target detection algorithms, and quantitative indicators to achieve intelligent monitoring and early warning of abnormal heat sources in the early stages of a fire. Hotspot regions are extracted from continuous infrared thermal images, and their spatial location stability, temperature change trends, thermal persistence, and spatial expansion characteristics are comprehensively analyzed. A fire confidence score is constructed to quantitatively assess the hotspots, effectively distinguishing between genuine fire sources and false hotspots caused by environmental interference or sensor noise. Simultaneously, the fire confidence score-based early warning triggering mechanism can respond promptly when hotspots show continuous temperature increases and significant spatial expansion, ensuring the timeliness and accuracy of early warnings and significantly reducing false alarm and missed alarm rates. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating an intelligent monitoring and early warning method for fire disasters based on infrared thermal imaging technology according to an embodiment of the present invention; Figure 2 This is a schematic illustration of a thermal infrared image of the area to be monitored according to an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the structural block diagram of a fire disaster intelligent monitoring and early warning system based on infrared thermal imaging technology according to an embodiment of the present invention. Detailed Implementation

[0023] Example of an intelligent monitoring and early warning method for fire disasters based on infrared thermal imaging technology.

[0024] like Figure 1 As shown in the flowchart, the intelligent monitoring and early warning method for fire disasters based on infrared thermal imaging technology according to an embodiment of the present invention includes the following steps: S1: Acquire multiple consecutive frames of thermal infrared images of the area to be monitored within a set time window.

[0025] In a preferred embodiment, an infrared thermal imager deployed in the area to be monitored is first used to continuously observe the area at a preset sampling frequency to obtain multiple frames of thermal infrared image data within a set time window. The thermal imager preferably has high thermal sensitivity and a wide dynamic range to ensure good response to slight temperature differences and rapid thermal changes, thereby guaranteeing the accuracy and real-time performance of subsequent image analysis results.

[0026] By acquiring data from consecutive frames within a set time window, the dynamic thermal characteristics of the monitored area can be effectively reflected, providing sufficient temporal information for identifying persistent hotspots and abnormal heating trends. Secondly, targeted image preprocessing operations are sequentially performed on the consecutive multi-frame thermal infrared images, including noise suppression and contrast enhancement.

[0027] Among them, noise suppression processing can employ wavelet threshold denoising algorithm based on spatial domain, anisotropic diffusion filtering algorithm, or nonlocal mean filtering algorithm to eliminate high-frequency pseudo-differences caused by sensor thermal noise, environmental interference, random noise, etc.

[0028] After noise suppression, to further enhance the distinguishability between hotspot and background regions in the image, contrast enhancement processing is performed on each frame of the thermal infrared image. Contrast enhancement can employ histogram equalization or multi-scale Retinex enhancement algorithms to expand the dynamic range of local temperature difference distribution.

[0029] S2: The adaptive threshold segmentation algorithm is used to process the continuous multi-frame thermal infrared images to obtain multiple hotspot regions, and the target hotspot region is obtained in the continuous multi-frame thermal infrared images using a target detection algorithm.

[0030] like Figure 2 As shown, this is a thermal infrared image of the area to be monitored in an embodiment of the present invention.

[0031] In a preferred embodiment, obtaining the target hotspot region in a series of consecutive thermal infrared images using a target detection algorithm specifically involves: calculating the IoU value between the target hotspot region and multiple hotspot regions in the series of consecutive thermal infrared images, and taking the hotspot region with the largest IoU value as the target hotspot region in the series of consecutive thermal infrared images.

[0032] For example: Obtain the target hotspot region in the first frame of thermal infrared image and all hotspot regions in the second frame of thermal infrared image; calculate the IoU value between the target hotspot region in the first frame of thermal infrared image and all hotspot regions in the second frame of thermal infrared image; and take the hotspot region with the largest IoU value as the target hotspot region in the second frame of thermal infrared image; in this example, the target hotspot region is any hotspot region in the first frame of thermal infrared image.

[0033] S3: Calculate the fire confidence level of the target hotspot area.

[0034] In a preferred embodiment, the first Fire confidence in key areas for: , For the first The extent of expansion in each hotspot area Using natural constants An exponential function with base 0. For the first The degree of fluctuation in each hotspot area This is the threshold for the degree of fluctuation.

[0035] By jointly modeling the spatial expansion characteristics and positional stability of hotspot areas, fire confidence can simultaneously reflect whether the heat source is continuously expanding and whether it possesses the low-fluctuation characteristics of a real fire point. Specifically, when a hotspot area exhibits significant expansion while maintaining a stable position, this indicator is kept at a high level, thereby enhancing the ability to identify early signs of real fires that are continuously heating up and expanding in shape. Conversely, when a hotspot experiences drastic fluctuations or random jumps in its spatial location, even with a certain expansion trend, the exponential decay mechanism rapidly weakens its confidence, effectively suppressing false hotspot alarms caused by factors such as sunlight reflection, mechanical equipment movement, and thermal noise flicker. Through coupled calculations of expansion and fluctuation behaviors, the system can more accurately distinguish between the formation process of real smoldering or open flames and hotspot interference caused by environmental noise, thus significantly improving the overall accuracy and robustness of early fire warnings.

[0036] No. The degree of fluctuation in each hotspot area for: , , For the first The hotspot area in the first Frame, First Centroid coordinates in a frame of thermal infrared image To set the total number of consecutive thermal infrared images within a time window, It is the Euclidean norm.

[0037] For example: , , The first The first frame of the thermal infrared image The first hotspot area The horizontal and vertical coordinates of each pixel For the first The first frame of the thermal infrared image The total number of pixels in each hotspot area.

[0038] By calculating the magnitude of the centroid trajectory change of each hotspot region over time in multiple consecutive frames of thermal infrared images, the spatial stability of the hotspot is quantified. Using the average coordinates of all pixels within the region as the centroid allows for a comprehensive reflection of the overall spatial distribution characteristics of the hotspot, rather than the random fluctuations of individual pixels, effectively avoiding the influence of noise points or local temperature disturbances. Since this fluctuation reflects the hotspot's positional shift across consecutive frames, it can clearly distinguish between real and relatively fixed continuous heat sources and false hotspots caused by sunlight reflection, equipment vibration, or background disturbances. This enables more accurate identification of abnormal heat sources, thereby improving the reliability and robustness of hotspot tracking, anomaly identification, and early warning.

[0039] No. Expansion of hotspot areas for: , For the first The duration of heat in each hotspot area To find the maximum value function, , The first The hotspot area in the first Frame, First The total number of pixels in a frame of thermal infrared image. To preset hyperparameters, This is used to set the total number of consecutive thermal infrared images within a time window.

[0040] By combining the area change trend of hotspot regions in continuous images with their temperature persistence, a more accurate characterization of real heating hotspots is achieved. On the one hand, by accumulating only the area growth portion and suppressing interference caused by area shrinkage, the indicator can highlight the expansion behavior of hotspots over time, thus effectively distinguishing the expansion of real heat sources caused by continuous heating from random fluctuations caused by noise, edge jitter, or illumination disturbances. On the other hand, thermal persistence is introduced as a regulating factor, ensuring that the expansion degree only significantly increases when the hotspot itself has stable heat release characteristics, avoiding misjudging short-lived or unstable pseudo-hotspots as abnormal areas. The expansion degree obtained in this way can more accurately capture the real physical evolution process of the heat source, improve the ability to identify early abnormal heating phenomena, and enhance the robustness of overall detection to complex environmental interference.

[0041] No. The duration of heat in each hotspot area for: , , These are the last frame and the first frame of the thermal infrared image within the set time window, respectively. Pixel temperature in hotspot areas , The first Frame, First The first frame of the thermal infrared image Pixel temperature in hotspot areas Using natural constants An exponential function with base 0. To set the total number of consecutive thermal infrared images within a time window, To define the first frame of multiple consecutive thermal infrared images within a set time window The linear regression slope of the pixel temperature change over time in each hotspot region. The pixel temperature of the hotspot area is the first The sum of the temperatures of all pixels within a hotspot area This is the preset first hyperparameter.

[0042] By comprehensively considering the overall temperature change amplitude, cumulative fluctuation, and temperature trend of hotspot areas over time, a thermal sustainability indicator is constructed that accurately reflects the stability and persistence of heat sources. Specifically, the ratio of the initial and final temperature differences to the overall fluctuation highlights whether the hotspot temperature is showing an overall increase or merely high-frequency fluctuations, better distinguishing between genuine heat sources with continuous temperature increases and unstable pseudo-hotspots. Simultaneously, an adjustment factor based on the time-series temperature fitting trend is introduced, significantly strengthening this indicator when the hotspot temperature shows a stable increase, while actively suppressing it when temperature changes are irregular or show a decaying trend. This design, which incorporates cumulative temperature changes, overall trends, and stability, accurately captures genuine continuous heat release behavior, thereby improving the reliability of early abnormal hotspot identification and effectively reducing misjudgments caused by environmental factors, instantaneous reflections, or noise fluctuations.

[0043] S4: Triggers an alert when the fire confidence level exceeds a preset threshold.

[0044] In a preferred embodiment, an early warning process is automatically triggered when the calculated fire confidence level of the hotspot area exceeds a preset threshold. This preset threshold can be adaptively set according to the fire sensitivity and false alarm tolerance of different scenarios to determine whether the hotspot area exhibits sufficiently significant and sustained abnormal temperature rise characteristics.

[0045] When an early warning is triggered, it not only immediately sends an alarm signal to the host computer or monitoring platform, but also simultaneously records the corresponding thermal infrared image, temperature distribution data, and hotspot evolution trajectory at the trigger moment, enabling traceable analysis of early fire signs. Furthermore, it can be linked with on-site audible and visual alarms, sprinkler systems, or intelligent inspection robots to proactively prevent fire risks from escalating. This significantly improves the accuracy, reliability, and response speed of early fire warnings, providing strong technical support for identifying early signs of fire.

[0046] This invention utilizes infrared thermal imaging technology for continuous dynamic observation of the monitored area. Combined with adaptive threshold segmentation and target detection algorithms, it accurately identifies and tracks hotspot areas. Furthermore, by comprehensively analyzing the temperature changes, spatial expansion, and location stability of these hotspots, it quantifies the fire confidence level, achieving early and accurate identification of potential fire sources. This not only captures real fire sources that are continuously heating and expanding spatially, but also effectively suppresses false alarms caused by environmental interference, sensor noise, or light reflection. Through a comprehensive assessment of thermal persistence, expansion degree, and fluctuation, the fire confidence level dynamically reflects the true risk of the fire source, significantly enhancing the robustness and accuracy of fire monitoring in complex environments. This enables intelligent, quantitative monitoring and reliable early warning of early fire signs.

[0047] Example of a fire disaster intelligent monitoring and early warning system based on infrared thermal imaging technology: like Figure 3 As shown in the figure, the structural block diagram of the intelligent fire disaster monitoring and early warning system based on infrared thermal imaging technology according to an embodiment of the present invention includes a processor and a memory.

[0048] This invention also provides an intelligent monitoring and early warning system for fire disasters based on infrared thermal imaging technology. For example... Figure 3 As shown, the system includes a processor and a memory. The memory stores computer program instructions. When the computer program instructions are executed by the processor, the intelligent monitoring and early warning method for fire disasters based on infrared thermal imaging technology according to the present invention is implemented.

[0049] The fire disaster intelligent monitoring and early warning system based on infrared thermal imaging technology also includes other components well known to those skilled in the art, such as communication interfaces. Their settings and functions are known in the art and will not be described in detail here.

[0050] In this invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used or combined with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store desired information and can be accessed by an application, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device. Any application or module described in this invention can be implemented using computer-readable / executable instructions stored or otherwise maintained by such a computer-readable medium.

[0051] In the description of this specification, "multiple" or "several" means at least two, such as two, three or more, unless otherwise explicitly specified.

[0052] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.

Claims

1. A method for intelligent monitoring and early warning of fire disasters based on infrared thermal imaging technology, characterized in that, include: Acquire multiple consecutive frames of thermal infrared images of the area to be monitored within a set time window; An adaptive threshold segmentation algorithm is used to process the consecutive multiple frames of thermal infrared images to obtain multiple hotspot regions, and a target detection algorithm is used to obtain the target hotspot region in the consecutive multiple frames of thermal infrared images; the target hotspot region is any hotspot region in the consecutive multiple frames of thermal infrared images. Calculate the fire confidence level of the target hotspot area and trigger an early warning when the fire confidence level exceeds a preset threshold; The fire confidence level is positively correlated with the expansion degree of the target hotspot area and negatively correlated with the fluctuation degree of the target hotspot area. The degree of expansion is positively correlated with the thermal persistence of the target hotspot area and the pixel change of the target hotspot area within a set time window; the thermal persistence is positively correlated with the pixel temperature of the target hotspot area in the last frame of the thermal infrared image within the set time window, and negatively correlated with the pixel temperature of the target hotspot area in the first frame of the thermal infrared image within the set time window and the linear regression slope of the pixel temperature of the target hotspot area in multiple consecutive frames of thermal infrared images within the set time window as a function of time.

2. The intelligent monitoring and early warning method for fire disasters based on infrared thermal imaging technology according to claim 1, characterized in that, No. Fire confidence in key areas for: , For the first The extent of expansion in each hotspot area Using natural constants An exponential function with base 0. For the first The degree of fluctuation in each hotspot area This is the threshold for the degree of fluctuation.

3. The intelligent monitoring and early warning method for fire disasters based on infrared thermal imaging technology according to claim 1, characterized in that, No. Expansion of hotspot areas for: , For the first The duration of heat in each hotspot area To find the maximum value function, , The first The hotspot area in the first Frame, First The total number of pixels in a frame of thermal infrared image. To preset hyperparameters, This is used to set the total number of consecutive thermal infrared images within a time window.

4. The intelligent monitoring and early warning method for fire disasters based on infrared thermal imaging technology according to claim 1, characterized in that, No. The duration of heat in each hotspot area for: , , These are the last frame and the first frame of the thermal infrared image within the set time window, respectively. Pixel temperature in hotspot areas , The first Frame, First The first frame of the thermal infrared image Pixel temperature in hotspot areas Using natural constants An exponential function with base 0. To set the total number of consecutive thermal infrared images within a time window, To define the first frame of multiple consecutive thermal infrared images within a set time window The slope of a linear regression of pixel temperature changes over time in hotspot regions. This is the preset first hyperparameter.

5. The intelligent monitoring and early warning method for fire disasters based on infrared thermal imaging technology according to claim 1, characterized in that, No. The degree of fluctuation in each hotspot area for: , , For the first The hotspot area in the first Frame, First Centroid coordinates in a frame of thermal infrared image To set the total number of consecutive thermal infrared images within a time window, It is the Euclidean norm.

6. The intelligent monitoring and early warning method for fire disasters based on infrared thermal imaging technology according to claim 1, characterized in that, Use an infrared thermal imager to acquire multiple consecutive frames of thermal infrared images of the area to be monitored within a set time window.

7. The intelligent monitoring and early warning method for fire disasters based on infrared thermal imaging technology according to claim 1, characterized in that, The specific steps for obtaining the target hotspot region in a series of consecutive thermal infrared images using a target detection algorithm are as follows: calculate the IoU value between the target hotspot region and multiple hotspot regions in the series of consecutive thermal infrared images, and take the hotspot region with the largest IoU value as the target hotspot region in the series of consecutive thermal infrared images.

8. The intelligent monitoring and early warning method for fire disasters based on infrared thermal imaging technology according to claim 1, characterized in that, The pixel temperature of the target hotspot area is the sum of the pixel temperatures within the target hotspot area.

9. The intelligent monitoring and early warning method for fire disasters based on infrared thermal imaging technology according to claim 1, characterized in that, The continuous multi-frame thermal infrared images are subjected to noise reduction and contrast enhancement processing.

10. A fire disaster intelligent monitoring and early warning system based on infrared thermal imaging technology, characterized in that, It includes a memory and a processor, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the intelligent monitoring and early warning method for fire disasters based on infrared thermal imaging technology as described in any one of claims 1 to 9 is implemented.

Images