A real-time fire prevention and extinguishing system for a belt conveyor system

Abstract

This invention relates to the field of industrial safety protection, and more particularly to a real-time fire prevention and extinguishing system for conveyor belt systems. The invention utilizes the collaborative operation of a data acquisition module, a pre-analysis module, a triggering module, a triggering module, and an early warning analysis module. Based on image acquisition units spaced along the conveyor belt path, the data acquisition module acquires images of the target area in real time. The pre-analysis module extracts environmental image parameters and identifies scattered feature blocks. The triggering module determines potential diffuse features and marks abnormal images based on the relative position and number of these scattered feature blocks. The triggering module periodically acquires images to construct an observation image set. The early warning analysis module determines the presence of anomalies through feature drift analysis. The fire extinguishing module activates fire extinguishing equipment deployed along the conveyor belt path, thereby improving the accuracy and timeliness of early warning of conveyor belt fires and ensuring the safe operation of the conveyor belt system.

Description

Technical Field

[0001] This invention relates to the field of industrial safety protection, and in particular to a real-time fire prevention and extinguishing system for belt conveyor systems. Background Technology

[0002] In modern industry, belt conveyors, as a highly efficient and continuous material transport method, are widely used in numerous industries such as mining, ports, power, and metallurgy. Taking the coal industry as an example, belt conveyors bear the crucial task of transporting coal from underground mining sites to surface coal preparation plants and subsequent transportation stages. Their transport capacity and stability directly affect the efficiency and cost of coal production. Statistics show that in large coal mining enterprises, belt conveyor systems account for over 80% of the total coal transport volume. In port logistics, belt conveyors are used for loading and unloading various bulk cargoes, such as ores and grains, greatly improving cargo throughput efficiency. Their efficient and continuous transport characteristics make production processes smoother, reduce material accumulation and transfer time, thereby lowering production costs and improving the economic benefits of enterprises.

[0003] Chinese Patent Publication No. CN119215362A discloses a fire prevention system and its control method, applied to a transfer station equipped with a belt conveyor. The system includes: a first detector, located inside the dust cover of the belt conveyor, for generating first fire information, indicating whether the conveyor belt is on fire; the conveyor belt is located inside the dust cover; a first fire extinguishing device, including a first water pipe located inside the dust cover and laid along the conveyor belt, with multiple spray holes distributed along the extension direction of the conveyor belt on the first water pipe; and a controller electrically connected to the first detector and the first fire extinguishing device, used to: acquire the first fire information from the first detector and control the first fire extinguishing device based on the first fire information to extinguish or cool the conveyor belt. This invention solves the technical problem of the high probability of fires in transfer stations.

[0004] However, the following problems still exist in the existing technology. In belt conveyor systems, combustibles such as coal, wood, and fibers form porous structures when piled up. This structure makes the combustibles more prone to smoldering when exposed to heat sources. The temperature change during the smoldering stage is not significant, and it is usually accompanied by dense smoke and a rise in temperature. However, in some transportation environments, dust distributed in the environment can mask the true characteristics of smoke, making it difficult for traditional infrared temperature detection and smoke detection methods to accurately capture early fire characteristics, easily leading to misjudgments and reducing the accuracy and timeliness of early fire warnings. Summary of the Invention

[0005] To address this issue, the present invention provides a real-time fire prevention and extinguishing system for belt conveyor systems. This system overcomes the problem in the prior art where, in some transportation environments, dust can obscure the true characteristics of smoke, making it difficult for traditional infrared temperature detection and smoke detection methods to accurately capture early fire characteristics, which can easily lead to misjudgments and reduce the accuracy and timeliness of early fire warnings.

[0006] To achieve the above objectives, the present invention provides a real-time fire prevention and extinguishing system for belt conveyor systems, comprising, The acquisition module includes image acquisition units that are spaced along the belt conveyor path to acquire images of the target area. The pre-analysis module is connected to the acquisition module to acquire images of each target area, extract environmental image parameters, and perform scattered feature recognition, including comparing the region blocks in the target area image with the environmental image parameters to identify scattered feature blocks in the target area image. The sensing trigger module is connected to the pre-analysis module. It determines potential diffuse features based on the relative positional relationship and number of scattered feature blocks to determine whether to mark the target area image and to determine the image acquisition unit corresponding to the target area image. The calling module, which is connected to the sensing triggering module, is used to determine the acquisition time corresponding to the target area image, and to acquire the target area image acquired by the image acquisition unit at a predetermined period interval, using the acquisition time as the starting time, to construct an observation image set. The early warning analysis module, which is connected to the calling module, is used to perform feature drift analysis based on the observed image set. It includes extracting scattered feature blocks from each of the target area images, constructing a main distribution area based on the relative positional relationship of the scattered feature blocks, determining the movement speed of the main distribution area based on the main distribution area in each target area image, and determining whether there is an anomaly by combining the movement speed of the conveyor belt. The fire extinguishing module, which responds to the early warning analysis module, is used to activate the fire extinguishing equipment deployed along the belt conveyor path.

[0007] Furthermore, the pre-analysis module is used to extract environmental image parameters, including: Used to extract the chromaticity of the target region image; The mean value of the chromaticity is used as an environmental image parameter.

[0008] Furthermore, the pre-analysis module is used to compare region blocks in the target region image with environmental image parameters to identify scattered feature blocks in the target region image, including... Used to segment the target region image into several region blocks; Used to extract the chromaticity of each of the aforementioned regions; Used to determine the difference ratio between the chromaticity of each of the said regions and the environmental image parameters; If the difference ratio is less than a preset threshold, the region block is identified as the scattered feature block.

[0009] Furthermore, the sensing triggering module is used to determine potential diffuse features based on the relative positional relationships and number of scattered feature blocks, including: Used to determine the average distance between each scattered feature block and its nearest scattered feature block; Used to determine the number of each scattered feature block; The reciprocal of the ratio of the average distance to the preset standard average distance is used as the first potential diffuse feature; The ratio of the stated quantity to a preset standard quantity is used as a second potential diffuse feature; The first potential diffuse feature and the second potential diffuse feature are weighted and summed to obtain the potential diffuse feature.

[0010] Furthermore, the sensing trigger module's method for determining whether to mark the target region image includes: If the potential diffuse feature is greater than or equal to a preset potential diffuse threshold, then the target region image is marked.

[0011] Furthermore, the calling module is also used to arrange the acquired target area images in the order of acquisition time to construct an observation image set.

[0012] Furthermore, the early warning analysis module is used to construct the main distribution area based on the relative positional relationships of scattered feature blocks, including: Used to determine the distances between scattered feature blocks; If the distance is less than a preset distance threshold, the scattered feature blocks are grouped together: Used to connect scattered feature blocks at the edges within the same group to construct the main distribution region.

[0013] Furthermore, the early warning analysis module is used to determine the movement speed of the main distribution area based on the main distribution area in the image of each target area, including: Used to determine the center of the main distribution region in each target region image; Used to calculate the displacement between the centers of each main distribution region in the observed image set; The ratio of each displacement to a predetermined periodic interval is used to determine the center moving speed; The average moving speed of each center is used to determine the moving speed of the main distribution area.

[0014] Furthermore, the early warning analysis module is used to determine whether any abnormalities exist, in conjunction with the conveyor belt's moving speed. Used to determine the speed difference ratio between the moving speed of the main distribution area and the moving speed of the conveyor belt; If the speed difference ratio is less than a preset speed difference ratio threshold, an anomaly is determined to exist.

[0015] Furthermore, the fire extinguishing module, used to activate fire extinguishing equipment deployed along the belt conveyor path, includes: If an anomaly is detected, the fire extinguishing equipment deployed along the belt conveyor path will be activated.

[0016] Compared with existing technologies, this invention improves the accuracy and timeliness of early warning for conveyor belt fires by establishing a collaborative system involving an acquisition module, a pre-analysis module, a triggering module, a triggering module, and an early warning analysis module. Based on image acquisition units spaced along the conveyor belt path, the acquisition module acquires images of the target area in real time. The pre-analysis module extracts environmental image parameters and identifies scattered feature blocks. The triggering module determines potential diffuse features and marks abnormal images based on the relative position and number of these scattered feature blocks. The triggering module periodically acquires images to construct an observation image set. The early warning analysis module determines the presence of anomalies through feature drift analysis. The fire suppression module activates fire suppression equipment deployed along the conveyor belt path. This enhances the accuracy and timeliness of early warning for conveyor belt fires, ensuring the safe operation of the conveyor belt system.

[0017] In particular, this invention identifies scattered feature blocks in the target area image through a pre-analysis module, and further determines potential diffuse features based on the relative positional relationship and quantity of the scattered feature blocks through a sensing trigger module, in order to determine whether to mark the target area image. In practice, the porous structure formed when transported objects are piled up during belt conveyor transport is prone to smoldering when it encounters a heat source. Early smoldering or fire often manifests as discrete features such as local smoke, high-temperature areas, or abnormal dust accumulation. These discrete features appear in the image as having significant color differences from the background environment. Therefore, the target area image can be segmented by the pre-analysis module to extract the color of each area block and compare it with the environmental image parameters, thereby identifying scattered feature blocks, which can provide a reliable basis for subsequent early warning. In the early stages of smoldering or fire, the diffusion of smoke, dust, or high-temperature areas is not completely uniform and continuous, but rather exhibits a dynamic process of evolution from point to surface and from discrete to aggregated. Furthermore, smoke or dust generated by a real fire typically has a continuous source and certain diffusion dynamics, which may cause scattered feature blocks to aggregate spatially, such as spreading around a point or along a certain direction. Conversely, accidental noise caused by instantaneous dust, equipment vibration, or light interference usually has scattered feature blocks that are spatially random and discrete. Therefore, by analyzing relative positional relationships, a large amount of spatially irrelevant random noise can be filtered out, capturing images of potentially anomaly-prone target areas. Simultaneously, the number of scattered feature blocks directly reflects the scale and intensity of potential anomalies. An isolated anomaly may originate from accidental interference, but the simultaneous occurrence of a large number of anomalies significantly increases the probability of a real hazard. Therefore, determining potential diffuse features from both relative positional relationships and quantity dimensions can more comprehensively and accurately identify early fire signs, providing a reliable basis for subsequent early warning analysis and firefighting decisions, thereby improving the accuracy and timeliness of early fire warnings.

[0018] In particular, this invention constructs an observation image set by acquiring images of the target area at predetermined periodic intervals using the image acquisition unit. In practice, image information at a single moment is easily affected by instantaneous environmental interference. For example, instantaneous dust storms may create color changes in the image that resemble early fire characteristics, potentially leading to misidentification as a potential anomaly. Therefore, by acquiring multiple images at predetermined periodic intervals and constructing an observation image set, the system can track and analyze the dynamic evolution of potentially anomaly-prone target area images over time. This continuous observation mechanism based on multiple frames not only reduces errors from single acquisitions and filters out occasional interference but also provides a data foundation for subsequent feature drift analysis by analyzing the movement trajectory and diffusion trend of scattered feature blocks in the sequence of images, thereby improving the accuracy and timeliness of early fire warnings.

[0019] In particular, this invention utilizes feature drift analysis based on observed image sets. In reality, the abnormal features of a real fire, such as smoke and high-temperature areas, originate from combustible materials on the conveyor belt. Their spatial movement follows the belt's transport pattern, meaning their apparent movement speed should be consistent with or highly correlated with the conveyor belt's operating speed. Static interference caused by non-fire factors such as changes in ambient light, equipment surface reflections, or instantaneous dust exhibits characteristics in the image sequence that appear randomly, remain stationary, or show movement patterns significantly inconsistent with belt transport. Therefore, by extracting scattered feature blocks from each time-series image and constructing a main distribution region, then calculating the movement speed of this main distribution region and comparing it with the system's known conveyor belt movement speed, a key criterion for distinguishing between real moving fires and static interference is established. This dynamic feature-based analysis method can reliably eliminate static or random pseudo-anomalies, ultimately achieving accurate identification and early warning of real early fires, thereby improving the accuracy and timeliness of early fire warnings. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the structure of a real-time fire prevention and extinguishing system for a belt conveyor system according to an embodiment of the invention. Figure 2 This is a logic decision diagram for identifying scattered feature blocks in an embodiment of the invention; Figure 3 This is a logic diagram for determining whether to mark a target region image according to an embodiment of the invention. Figure 4 This is a logic diagram for determining whether an anomaly exists, as shown in an embodiment of the invention. Detailed Implementation

[0021] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0022] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0023] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the term "connection" should be interpreted broadly. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0024] Please see Figure 1 The diagram shown is a structural schematic of a real-time fire prevention and extinguishing system for a belt conveyor system according to an embodiment of the invention. The real-time fire prevention and extinguishing system for a belt conveyor system of the present invention includes: The acquisition module includes image acquisition units that are spaced along the belt conveyor path to acquire images of the target area. The pre-analysis module is connected to the acquisition module to acquire images of each target area, extract environmental image parameters, and perform scattered feature recognition, including comparing the region blocks in the target area image with the environmental image parameters to identify scattered feature blocks in the target area image. The sensing trigger module is connected to the pre-analysis module. It determines potential diffuse features based on the relative positional relationship and number of scattered feature blocks to determine whether to mark the target area image and to determine the image acquisition unit corresponding to the target area image. The calling module, which is connected to the sensing triggering module, is used to determine the acquisition time corresponding to the target area image, and to acquire the target area image acquired by the image acquisition unit at a predetermined period interval, using the acquisition time as the starting time, to construct an observation image set. The early warning analysis module, which is connected to the calling module, is used to perform feature drift analysis based on the observed image set. It includes extracting scattered feature blocks from each of the target area images, constructing a main distribution area based on the relative positional relationship of the scattered feature blocks, determining the movement speed of the main distribution area based on the main distribution area in each target area image, and determining whether there is an anomaly by combining the movement speed of the conveyor belt. The fire extinguishing module, which responds to the early warning analysis module, is used to activate the fire extinguishing equipment deployed along the belt conveyor path.

[0025] Specifically, there are no restrictions on the specific structure of the pre-analysis module, the sensing trigger module, the calling module, the early warning analysis module, and the fire extinguishing module. They can all be composed of logic components, including field-programmable processors, computers, or microprocessors in computers.

[0026] Specifically, there are no restrictions on the specific structure of the acquisition module. The image acquisition unit can be an industrial photography device, as long as it can acquire images of the monitoring area.

[0027] Specifically, there are no restrictions on the form of the fire extinguishing equipment. For example, it can be a sprinkler system that is remotely controlled and activated by the fire extinguishing module.

[0028] Specifically, the pre-analysis module is used to extract environmental image parameters, including: Used to extract the chromaticity of the target region image; The mean value of the chromaticity is used as an environmental image parameter.

[0029] Specifically, there is no limitation on the method of extracting the chromaticity of the target area image. The chromaticity value of each pixel in the image can be directly calculated by image processing technology, or chromaticity extraction can be performed using professional image analysis software. As long as it can accurately reflect the chromaticity of the target area image, it will not be elaborated further.

[0030] Please see Figure 2 As shown, this is a logical decision diagram for identifying scattered feature blocks according to an embodiment of the invention. Specifically, the pre-analysis module is used to compare the region blocks in the target region image with environmental image parameters to identify scattered feature blocks in the target region image, including... Used to segment the target region image into several region blocks; Used to extract the chromaticity of each of the aforementioned regions; Used to determine the difference ratio between the chromaticity of each of the said regions and the environmental image parameters; If the difference ratio is less than a preset threshold, the region block is identified as the scattered feature block.

[0031] In practice, there is no limitation on the method of image segmentation for the target region image. It can be done by regular grid division or other methods, as long as the image can be systematically divided into several regions that can be analyzed independently. This will not be elaborated further.

[0032] In practice, the purpose of the preset threshold is to characterize the degree of difference between the color intensity of the area block and the environmental background. The threshold is predetermined, and it is usually selected within the range of [0.3, 0.5]. In practice, it is preferred to be 0.4.

[0033] The difference ratio is the ratio of the absolute value of the difference between two values ​​to the mean of the two values.

[0034] Specifically, the sensing triggering module is used to determine potential diffuse features based on the relative positional relationships and number of scattered feature blocks, including: Used to determine the average distance between each scattered feature block and its nearest scattered feature block; Used to determine the number of each scattered feature block; The reciprocal of the ratio of the average distance to the preset standard average distance is used as the first potential diffuse feature; The ratio of the stated quantity to a preset standard quantity is used as a second potential diffuse feature; The first potential diffuse feature and the second potential diffuse feature are weighted and summed to obtain the potential diffuse feature.

[0035] In practice, there is no limitation on the method of determining the average distance between each scattered feature block and its nearest scattered feature block. The center of the image block can be determined by image processing technology, and then the distance from each feature block to its nearest neighbor can be calculated and the average of all distances can be obtained. This will not be elaborated further.

[0036] In implementation, the purpose of the preset standard average distance is to standardize the distance distribution between scattered feature blocks in order to provide a benchmark reference for the determination of potential diffuse features. The standard average distance is predetermined. Those skilled in the art can determine the average distance between scattered feature blocks under normal conditions by statistically analyzing a large number of image samples under historical conditions with smoke. The standard average distance is set as the product of the average distance and the distance offset coefficient. Typically, the distance offset coefficient is selected within the range of [1.15, 1.35], and is preferably 1.15 in implementation.

[0037] In implementation, the purpose of the preset standard quantity is to characterize the reasonable range of the number of scattered feature blocks under normal operating conditions, providing a quantitative basis for determining whether there are potential anomalies. The standard quantity is predetermined. Those skilled in the art can determine the average number of scattered feature blocks under normal conditions by statistically analyzing a large number of image samples from historical operating conditions with smoke, thus representing the distribution of scattered feature blocks under abnormal conditions. To represent possible fluctuations in the number, the preset standard quantity is set as the product of the average number and the number offset coefficient. Typically, the number offset coefficient is selected within the range of [0.65, 0.95], and is preferably 0.85 in implementation.

[0038] In practice, when the first potential diffuse feature and the second potential diffuse feature are weighted and summed, the weight of the first potential diffuse feature is 0.55 and the weight of the second potential diffuse feature is 0.45.

[0039] Please see Figure 3 As shown, this is a logic determination diagram for determining whether to mark a target region image according to an embodiment of the invention. Specifically, the sensing trigger module is used to determine whether to mark the target region image, including... If the potential diffuse feature is greater than or equal to a preset potential diffuse threshold, then the target region image is marked.

[0040] In practice, the purpose of the preset potential diffusion threshold is to characterize the abnormal distribution of scattered feature blocks in the current target area image and whether there is a fire hazard. Under normal circumstances, the potential diffusion threshold is selected in the range [0.95, 1.25], and in practice, it is preferably 1.15.

[0041] This invention identifies scattered feature blocks in the target area image through a pre-analysis module, and further determines potential diffuse features based on the relative position and quantity of the scattered feature blocks through a sensing trigger module, in order to determine whether to mark the target area image. In practice, the porous structure formed when transported objects are piled up during belt conveyor transport is prone to smoldering when it encounters a heat source. Early smoldering or fire often manifests as discrete features such as local smoke, high-temperature areas, or abnormal dust accumulation. These discrete features appear in the image as having significant color differences from the background environment. Therefore, the target area image can be segmented by the pre-analysis module to extract the color of each area block and compare it with the environmental image parameters, thereby identifying scattered feature blocks, which can provide a reliable basis for subsequent early warning. In the early stages of smoldering or fire, the diffusion of smoke, dust, or high-temperature areas is not completely uniform and continuous, but rather exhibits a dynamic process of evolution from point to surface and from discrete to aggregated. Furthermore, smoke or dust generated by a real fire typically has a continuous source and certain diffusion dynamics, which may cause scattered feature blocks to aggregate spatially, such as spreading around a point or along a certain direction. Conversely, accidental noise caused by instantaneous dust, equipment vibration, or light interference usually has scattered feature blocks that are spatially random and discrete. Therefore, by analyzing relative positional relationships, a large amount of spatially irrelevant random noise can be filtered out, capturing images of potentially anomaly-prone target areas. Simultaneously, the number of scattered feature blocks directly reflects the scale and intensity of potential anomalies. An isolated anomaly may originate from accidental interference, but the simultaneous occurrence of a large number of anomalies significantly increases the probability of a real hazard. Therefore, determining potential diffuse features from both relative positional relationships and quantity dimensions can more comprehensively and accurately identify early fire signs, providing a reliable basis for subsequent early warning analysis and firefighting decisions, thereby improving the accuracy and timeliness of early fire warnings.

[0042] Specifically, the calling module is also used to arrange the acquired target area images in the order of acquisition time to construct an observation image set.

[0043] In implementation, the calling module acquires images of the target area collected by the image acquisition unit at predetermined periodic intervals, and arranges these images sequentially according to the order of acquisition time to form an ordered set of observation images, ensuring the temporal consistency of the image sequence, so as to provide accurate time dimension data for subsequent feature drift analysis. The predetermined periodic interval is predetermined, and is usually selected within the range [0.3s, 1s], preferably 0.5s in implementation.

[0044] This invention constructs an observation image set by acquiring images of the target area at predetermined periodic intervals using an image acquisition unit. In practice, image information at a single moment is easily affected by instantaneous environmental interference. For example, instantaneous dust storms may create color changes in the image that resemble early fire characteristics, potentially leading to misidentification as a potential anomaly. Therefore, by acquiring multiple images at predetermined periodic intervals and constructing an observation image set, the system can track and analyze the dynamic evolution of potentially anomaly-prone target area images over time. This continuous observation mechanism based on multiple frames not only reduces errors from single acquisitions and filters out occasional interference but also provides a data foundation for subsequent feature drift analysis by analyzing the movement trajectory and diffusion trend of scattered feature blocks in the sequence of images, thereby improving the accuracy and timeliness of early fire warnings.

[0045] Specifically, the early warning analysis module is used to construct the main distribution area based on the relative positional relationships of scattered feature blocks, including: Used to determine the distances between scattered feature blocks; If the distance is less than a preset distance threshold, the scattered feature blocks are grouped together: Used to connect scattered feature blocks at the edges within the same group to construct the main distribution region.

[0046] In implementation, the purpose of the preset distance threshold is to characterize the maximum allowable spacing between scattered feature blocks. The distance threshold is predetermined. Those skilled in the art can statistically analyze the abnormal feature blocks in image samples under historical smoke conditions to obtain the average distance between scattered feature blocks. This average distance represents the spatial density that feature points within a continuous abnormal area should have under normal conditions. To represent an early fire, a more rigorous and conservative clustering criterion is used to avoid incorrectly including spatially irrelevant discrete noise points in the main distribution area. The distance threshold is set as the product of the average distance and the distance error coefficient. Typically, the distance error coefficient is selected within the range of [0.65, 0.85], and is preferably 0.65 in implementation.

[0047] In practice, there are no restrictions on the method for determining the geometric center of the main distribution area. Those skilled in the art can choose any existing means of determining the geometric center of irregular shapes, which will not be elaborated here.

[0048] Specifically, the early warning analysis module is used to determine the movement speed of the main distribution area based on the main distribution area in the image of each target area, including: Used to determine the center of the main distribution region in each target region image; Used to calculate the displacement between the centers of each main distribution region in the observed image set; The ratio of each displacement to a predetermined periodic interval is used to determine the center moving speed; The average moving speed of each center is used to determine the moving speed of the main distribution area.

[0049] In practice, there is no limitation on the method of calculating the displacement between the centers of each main distribution region in the observed image set. The displacement can be determined by calculating the pixel coordinate difference between the center points of the main distribution regions in two adjacent frames of continuous time-series images, or by using image registration technology. As long as the displacement between the centers of each main distribution region can be accurately calculated, this will not be elaborated further.

[0050] Please see Figure 4 As shown, this is a logic diagram for determining whether an anomaly exists according to an embodiment of the invention. Specifically, the early warning analysis module is used to determine whether an anomaly exists by combining the conveyor belt's moving speed, including... Used to determine the speed difference ratio between the moving speed of the main distribution area and the moving speed of the conveyor belt; If the speed difference ratio is less than a preset speed difference ratio threshold, an anomaly is determined to exist.

[0051] In practice, the purpose of the preset speed difference ratio threshold is to define a maximum allowable speed deviation tolerance for determining whether the moving speed of the main distribution area and the moving speed of the conveyor belt are synchronized. In the case of small differences between the pointers, the speed difference ratio threshold is selected in the range [0.15, 0.35], preferably 0.25.

[0052] Specifically, the fire extinguishing module is used to activate fire extinguishing equipment deployed along the belt conveyor path, including: If an anomaly is detected, the fire extinguishing equipment deployed along the belt conveyor path will be activated.

[0053] This invention utilizes feature drift analysis based on observed image sets. In reality, the abnormal features of a real fire, such as smoke and high-temperature areas, originate from flammable materials on a conveyor belt. Their spatial movement follows the belt's transport pattern, meaning their apparent movement speed should be consistent with or highly correlated with the conveyor belt's operating speed. Static interference caused by non-fire factors such as changes in ambient light, equipment surface reflections, or instantaneous dust exhibits characteristics in the image sequence that appear randomly, remain stationary, or show movement patterns significantly inconsistent with belt transport. Therefore, by extracting scattered feature blocks from each time-series image and constructing a main distribution region, then calculating the movement speed of this main distribution region and comparing it with the system's known conveyor belt movement speed, a key criterion for distinguishing between real moving fires and static interference is established. This dynamic feature-based analysis method can reliably eliminate static or random pseudo-anomalies, ultimately achieving accurate identification and early warning of real early fires, thereby improving the accuracy and timeliness of early fire warnings.

[0054] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A real-time fire prevention and extinguishing system for belt conveyor systems, characterized in that, include, The acquisition module includes image acquisition units that are spaced along the belt conveyor path to acquire images of the target area. The pre-analysis module is connected to the acquisition module to acquire images of each target area, extract environmental image parameters, and perform scattered feature recognition, including comparing the region blocks in the target area image with the environmental image parameters to identify scattered feature blocks in the target area image. A sensing trigger module, connected to the pre-analysis module, determines potential diffuse features based on the relative positional relationship and quantity of scattered feature blocks to determine whether to mark the target region image and identify the image acquisition unit corresponding to the target region image; wherein, determining the potential diffuse features includes, Used to determine the average distance between each scattered feature block and its nearest scattered feature block; Used to determine the number of each scattered feature block; The reciprocal of the ratio of the average distance to the preset standard average distance is used as the first potential diffuse feature; The ratio of the stated quantity to a preset standard quantity is used as a second potential diffuse feature; The first potential diffuse feature and the second potential diffuse feature are weighted and summed to obtain the potential diffuse feature; The calling module, which is connected to the sensing triggering module, is used to determine the acquisition time corresponding to the target area image, and to acquire the target area image acquired by the image acquisition unit at a predetermined period interval, using the acquisition time as the starting time, to construct an observation image set. The early warning analysis module, which is connected to the calling module, is used to perform feature drift analysis based on the observed image set. It includes extracting scattered feature blocks from each of the target area images, constructing a main distribution area based on the relative positional relationship of the scattered feature blocks, determining the movement speed of the main distribution area based on the main distribution area in each target area image, and determining whether there is an anomaly by combining the movement speed of the conveyor belt. The fire extinguishing module, which responds to the early warning analysis module, is used to activate the fire extinguishing equipment deployed along the belt conveyor path.

2. The real-time fire prevention and extinguishing system for belt conveyor systems according to claim 1, characterized in that, The pre-analysis module is used to extract environmental image parameters, including... Used to extract the chromaticity of the target region image; The mean value of the chromaticity is used as an environmental image parameter.

3. The real-time fire prevention and extinguishing system for belt conveyor systems according to claim 1, characterized in that, The pre-analysis module is used to compare region blocks in the target region image with environmental image parameters to identify scattered feature blocks in the target region image, including... Used to segment the target region image into several region blocks; Used to extract the chromaticity of each of the aforementioned regions; Used to determine the difference ratio between the chromaticity of each of the said regions and the environmental image parameters; If the difference ratio is less than a preset threshold, the region block is identified as the scattered feature block.

4. The real-time fire prevention and extinguishing system for belt conveyor systems according to claim 3, characterized in that, The sensing trigger module is used to determine whether to mark the target area image, including... If the potential diffuse feature is greater than or equal to a preset potential diffuse threshold, then the target region image is marked.

5. The real-time fire prevention and extinguishing system for belt conveyor systems according to claim 1, characterized in that, The calling module is also used to arrange the acquired target area images in the order of acquisition time to construct an observation image set.

6. The real-time fire prevention and extinguishing system for belt conveyor systems according to claim 1, characterized in that, The early warning analysis module is used to construct the main distribution area based on the relative positional relationships of scattered feature blocks, including: Used to determine the distances between scattered feature blocks; If the distance is less than a preset distance threshold, the scattered feature blocks are grouped together: Used to connect scattered feature blocks at the edges within the same group to construct the main distribution region.

7. The real-time fire prevention and extinguishing system for belt conveyor systems according to claim 1, characterized in that, The early warning analysis module is used to determine the movement speed of the main distribution area based on the main distribution area in the image of each target area, including: Used to determine the center of the main distribution region in each target region image; Used to calculate the displacement between the centers of each main distribution region in the observed image set; The ratio of each displacement to a predetermined periodic interval is used to determine the center moving speed; The average moving speed of each center is used to determine the moving speed of the main distribution area.

8. The real-time fire prevention and extinguishing system for belt conveyor systems according to claim 7, characterized in that, The early warning analysis module is used to determine whether there are any abnormalities by combining the conveyor belt's moving speed. Used to determine the speed difference ratio between the moving speed of the main distribution area and the moving speed of the conveyor belt; If the speed difference ratio is less than a preset speed difference ratio threshold, an anomaly is determined to exist.

9. The real-time fire prevention and extinguishing system for belt conveyor systems according to claim 8, characterized in that, The fire extinguishing module is used to activate the fire extinguishing equipment deployed along the belt conveyor path, including: If an anomaly is detected, the fire extinguishing equipment deployed along the belt conveyor path will be activated.

Images