Intelligent fire-fighting fire smoke particle detection method

Abstract

The application relates to the technical field of intelligent sensors, in particular to a fire smoke particle detection method for intelligent fire fighting, which comprises the following steps: obtaining multiple target images of a target area, determining whether a smoke area in each target image is a fire smoke area according to the gray value of the smoke area in each target image; in the case of the fire smoke area, determining the actual brightness of each sub-area in the fire smoke area; determining the global concentration of the fire smoke area according to the original brightness, the actual brightness and the pixel number of each sub-area, and determining the instantaneous increase rate of the smoke concentration based on the global concentration of the fire smoke area in the target image of the current frame and the global concentration of the fire smoke area in the target image of the last frame; and determining the fire risk level by using the instantaneous increase rate of the smoke concentration and triggering the alarm of the corresponding level. Thus, the application improves the accuracy and reliability of the detection result of the fire smoke particle.

Description

Technical Field

[0001] This invention relates to the field of intelligent sensor technology, and specifically to a method for detecting smoke particles in intelligent fire protection. Background Technology

[0002] Fire, as a destructive disaster, constantly threatens people's lives and property. According to statistics released by the National Fire and Rescue Administration, fire accidents have occurred frequently in my country in recent years, causing significant casualties and economic losses. Fire smoke, as a core physical characteristic for early fire warning, plays a decisive role in the effectiveness of the entire fire warning system due to its timeliness and accuracy of detection.

[0003] In some scenarios, smoke particle detection in fires is often based on a single sensing principle. For example, ionization smoke sensors detect fires by detecting the effect of smoke particles on ion current, while photoelectric smoke sensors trigger alarms by utilizing the scattering or blocking of light by smoke. However, these technologies exhibit serious drawbacks in complex environments, such as high false alarm rates and insufficient reliability, often leading to sensor misjudgments and unnecessary alarms. Therefore, using these methods for fire smoke particle detection can easily result in low accuracy and reliability of the detection results. Summary of the Invention

[0004] To address the technical problem of low accuracy and reliability in detecting smoke particles during fires, the present invention aims to provide an intelligent fire protection method for detecting smoke particles. The specific technical solution adopted is as follows:

[0005] This invention provides a method for detecting smoke particles in intelligent fire protection, comprising: acquiring multiple frames of target images of a target area; determining whether the smoke area in each frame of the target image is a fire smoke area based on the grayscale value of the smoke area; if it is a fire smoke area, determining the actual brightness of each sub-region in the fire smoke area based on the original brightness, brightness attenuation coefficient, and distance of the light source center in the fire smoke area when there is no smoke, wherein the distance is the distance between the position of each sampling point set along the optical axis of the light source center and the light source center, and the sub-region is formed by segmenting the fire smoke area by each sampling point set along the optical axis of the light source center; determining the global concentration of the fire smoke area based on the original brightness, the actual brightness of each sub-region, and the number of pixels, and determining the instantaneous increase rate of smoke concentration based on the global concentration of the fire smoke area in the target image of the current frame and the global concentration of the fire smoke area in the target image of the previous frame; and determining the fire risk level and triggering an alarm of the corresponding level using the instantaneous increase rate of smoke concentration.

[0006] Optionally, acquiring multiple frames of target images of the target area includes: comparing the image of the target area captured by the camera with the background image, and calculating the pixel-level difference between the current frame image and the background image using the image difference method; if the pixel-level difference exceeds a first threshold, determining that the target area is suspected of having a fire; and capturing the target area at high frame rates with the camera to obtain the target image, wherein the frame rate of the high frame rate capture is the ratio between the smoke diffusion speed and the maximum allowable displacement per frame.

[0007] Optionally, determining whether a smoke region in each frame of the target image is a fire smoke region based on the grayscale value of the smoke region in each frame of the target image includes: segmenting the smoke region from each target image based on the grayscale difference between each frame of the target image and the background image; determining the area change trend based on the area of ​​the core region of the smoke region in the target images of adjacent frames, and determining the grayscale change trend based on the average grayscale of the core region of the smoke region in the target images of adjacent frames, wherein the core region is the high grayscale core region where smoke particles are concentrated in the smoke region; determining the smoke region as a suspected fire smoke region if both the area change trend and the grayscale change trend are greater than the second threshold; in the case of a suspected fire smoke region, selecting feature pixels with rich texture from the suspected fire smoke region; determining the angle between the motion vector of the feature pixel and the unit vector in the vertical direction based on the horizontal and vertical components of the feature pixel, wherein the motion vector is determined by comparing the positional changes of the same feature pixel in the target images of adjacent frames; and determining the suspected fire smoke region as a fire smoke region if the proportion of feature pixels with an angle less than the third threshold is greater than or equal to the fourth threshold.

[0008] Optionally, determining the area change trend based on the area of ​​the core region of the smoke region in the target image of adjacent frames includes: calculating a first difference between the area of ​​the core region of the smoke region in the next frame of the target image and the area of ​​the core region of the smoke region in the current frame; determining a first ratio between the first difference and the area of ​​the core region of the smoke region in the current frame as the area change trend of the core region of the smoke region in the target image of the current frame; determining the grayscale change trend based on the average grayscale of the core region of the smoke region in the target image of adjacent frames includes: calculating a second difference between the average grayscale of the core region of the smoke region in the next frame of the target image and the average grayscale of the core region of the smoke region in the current frame; determining a second ratio between the second difference and the average grayscale of the core region of the smoke region in the current frame as the grayscale change trend of the core region of the smoke region in the target image of the current frame.

[0009] Optionally, determining the angle between the motion vector of the feature pixel and the unit vector in the vertical direction based on the horizontal and vertical components of the feature pixel includes: calculating the square root of the sum of the squares of the horizontal and vertical components, and calculating the third ratio between the vertical component and the square root of the sum; processing the third ratio using the inverse cosine function to obtain the angle.

[0010] Optionally, the actual brightness of each sub-region in the fire smoke area is determined based on the original brightness, brightness attenuation coefficient, and distance of the light source center in the smoke-free area. This includes: calculating the first product between the absolute value of the brightness attenuation coefficient and the distance; and determining the third difference between the original brightness and the first product as the actual brightness.

[0011] Optionally, determining the global concentration of the fire smoke region based on the original brightness, the actual brightness of each sub-region, and the number of pixels includes: determining the smoke concentration of each sub-region based on the actual brightness, calibration constant, and original brightness of each sub-region in the fire smoke region; calculating the second product between the smoke concentration of each sub-region and the corresponding number of pixels, and superimposing the second products to obtain a first superimposed value; superimposing the number of pixels of each sub-region to obtain a second superimposed value; and determining the fourth ratio between the first superimposed value and the second superimposed value as the global concentration of the fire smoke region.

[0012] Optionally, determining the smoke concentration of each sub-region based on the actual brightness, calibration constant, and original brightness of each sub-region in the fire smoke area includes: calculating the average brightness of the actual brightness of each sub-region in the fire smoke area, and calculating a fifth ratio between the average brightness and the original brightness; calculating a fourth difference between a predetermined value and the fifth ratio; and determining the third product between the calibration constant and the fourth difference as the smoke concentration.

[0013] Optionally, determining the instantaneous rate of increase of smoke concentration based on the global concentration of the fire smoke region in the target image of the current frame and the global concentration of the fire smoke region in the target image of the previous frame includes: calculating the fifth difference between the global concentration of the fire smoke region in the target image of the current frame and the global concentration of the fire smoke region in the target image of the previous frame, and the time interval between the target image of the current frame and the target image of the previous frame; and determining the sixth ratio between the fifth difference and the time interval as the instantaneous rate of increase.

[0014] Optionally, determining the fire risk level and triggering the corresponding alarm based on the instantaneous increase in smoke concentration includes: determining the fire risk level as primary and triggering a primary alarm when the instantaneous increase in smoke concentration is greater than the fifth threshold and less than the sixth threshold, and notifying staff to check when a primary alarm is triggered; determining the fire risk level as intermediate and triggering a intermediate alarm when the instantaneous increase in smoke concentration is greater than or equal to the sixth threshold and less than the seventh threshold, and shutting down the ventilation system when a intermediate alarm is triggered; and determining the fire risk level as high and triggering a high alarm when the instantaneous increase in smoke concentration is greater than or equal to the seventh threshold, and activating sprinklers when a high alarm is triggered.

[0015] The present invention has the following beneficial effects: First, multiple frames of target images of the target area are acquired, and the grayscale value of the smoke area in each frame of the target image is used to determine whether the smoke area in each frame of the target image is a fire smoke area; then, if it is a fire smoke area, the actual brightness of each sub-region in the fire smoke area is determined based on the original brightness, brightness attenuation coefficient, and distance of the light source center in the fire smoke area when there is no smoke. The distance is the distance between the position of each sampling point set along the optical axis of the light source center and the light source center, and the sub-region is formed by segmenting the fire smoke area by each sampling point set along the optical axis of the light source center; second, the global concentration of the fire smoke area is determined based on the original brightness, the actual brightness of each sub-region, and the number of pixels, and the instantaneous increase rate of smoke concentration is determined based on the global concentration of the fire smoke area in the target image of the current frame and the global concentration of the fire smoke area in the target image of the previous frame; finally, the instantaneous increase rate of smoke concentration is used to determine the fire risk level and trigger the corresponding alarm level.

[0016] Thus, this invention enables the non-contact detection of fire smoke particles by photographing a target area. Based on differences in grayscale values ​​and brightness within the captured images, the fire smoke situation within the target area is determined. By combining multiple captured images, a fire risk level for the fire smoke is constructed, triggering a corresponding alarm. Therefore, compared to using a single sensor, the method provided by this invention is not affected by complex environments, improving the accuracy and reliability of fire smoke particle detection results. Attached Figure Description

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

[0018] Figure 1A flowchart of a fire smoke particle detection method for intelligent fire protection provided in an embodiment of the present invention;

[0019] Figure 2 This is a schematic diagram of the layout of an image capturing system provided in an embodiment of the present invention;

[0020] Figure 3 A schematic diagram illustrating the changing trend of a core region provided in an embodiment of the present invention;

[0021] Figure 4 This is a schematic diagram illustrating the distribution of sub-regions within a fire smoke area, provided as an embodiment of the present invention.

[0022] Figure 5 This is a schematic diagram of an attenuation model provided in an embodiment of the present invention. Detailed Implementation

[0023] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an intelligent fire-fighting smoke particle detection method proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0025] The specific scheme of the intelligent fire protection smoke particle detection method provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0026] Example 1:

[0027] Please see Figure 1 The diagram illustrates a flowchart of a fire smoke particle detection method for intelligent fire protection according to an embodiment of the present invention, comprising:

[0028] Step S101: Obtain multiple frames of target images of the target area, and determine whether the smoke area in each frame of the target image is a fire smoke area based on the gray value of the smoke area in each frame of the target image.

[0029] Specifically, this embodiment of the invention achieves data acquisition under stable lighting conditions throughout the day by deploying a constant-light LED light source and combining it with global monitoring. When suspected fire smoke (such as smoke characteristics generated in the early stages of a fire) is detected in the target area, a high-frame-rate shooting mode is automatically triggered to provide clear and continuous image data support for subsequent fire situation assessment. The target area can be a warehouse, etc. For example, such as... Figure 2 As shown, Figure 2 This is a schematic diagram of the layout of an image capturing system provided in an embodiment of the present invention. Figure 2 In this system, a constant-light LED source is positioned at the top of the target area, covering the entire area and allowing for omnidirectional image capture. Upon system startup, initialization occurs during periods when no personnel or goods are present in the target area (e.g., late at night), capturing dozens of static photos including the goods to generate an "average background image." This image records the brightness, fixed structure outlines, and texture of the warehouse under normal conditions, serving as a baseline for subsequent detection. After initialization, real-time monitoring begins. During this phase, the camera continuously captures video, comparing each frame with the background image. Image differencing is used to calculate pixel-level differences between the current frame and the background image, detecting dynamic changes in the target area. Since smoke areas obstruct light, their grayscale values ​​are lower than the background values. If a frame shows a localized decrease in brightness, it is marked as a suspected fire / smoke area, and high-frame-rate recording is initiated to record the smoke's behavior.

[0030] Furthermore, as an optional embodiment of the present invention, acquiring multiple frames of target images of the target area includes: comparing the image of the target area captured by the camera with the background image, and calculating the pixel-level difference between the current frame image and the background image using the image difference method; if the pixel-level difference exceeds a first threshold, determining that a fire is suspected to have occurred in the target area; and capturing the target area at high frame rates with the camera to obtain the target image, wherein the frame rate of the high frame rate capture is the ratio between the smoke diffusion speed and the maximum allowable displacement per frame.

[0031] Specifically, in this embodiment of the invention, the pixel-level difference between the current frame and the background image is calculated using image differencing to detect dynamic changes in the target area. Because smoke areas obstruct light, the grayscale value in the current frame is lower than that in the background image. If a frame shows a localized decrease in brightness, it is marked as a suspected area of ​​fire smoke, and high-frame-rate recording is initiated to record the smoke's behavior. The first threshold can be determined based on actual conditions, and this embodiment of the invention does not impose a limitation. The method for determining pixel-level differences between different images using image differencing can refer to known techniques, which will not be elaborated upon here.

[0032] Furthermore, when determining the frame rate for high-frame-rate shooting, the following formula can be used for calculation:

[0033]

[0034] In the above formula, This indicates the frame rate of high-frame-rate shooting. This represents the smoke diffusion speed, typically ranging from 0.5 to 2 m / s. This represents the maximum allowable displacement per frame, used to ensure the continuity of the smoke's trajectory. A value range of 0.01~0.03m is recommended.

[0035] Furthermore, for multi-frame target images acquired from high frame rate image sequences, to achieve accurate identification of fire smoke areas, the first step is to analyze the grayscale difference between the smoke area and the background to accurately segment the smoke area in a single frame target image. For example, neural networks can be used to achieve accurate smoke area segmentation, providing basic visual information for subsequent fire identification. Specifically, in a warehouse with fixed light sources, the smoke-covered area exhibits semi-transparent characteristics due to decreased light transmittance. Its grayscale value lies between the completely obscured shadow area and the directly illuminated area, resulting in a concentrated distribution of grayscale values ​​in the mid-gray range, exhibiting a single-peak or double-peak distribution (split into a main peak and a diffusion peak when the concentration is uneven). Based on this pattern, an initial threshold range is manually set by analyzing the peak and valley positions of the histogram, and the threshold boundary is optimized using the maximum inter-class variance algorithm to improve segmentation accuracy. Finally, a 5×5 elliptic kernel closure operation is used to smooth the burrs or breaks at the edges of the smoke clumps, taking into account the clump-like diffusion characteristics of the smoke.

[0036] Furthermore, in this embodiment of the invention, to ensure the accuracy and reliability of fire smoke area determination, the embodiment also uses temporal characteristic analysis of high frame rate image sequences. By capturing the diffusion behavior of smoke in the spatiotemporal domain and combining it with the unique motion characteristics of fire smoke, a hybrid determination strategy is adopted. This involves analyzing the changes in the region (smoke concentration and motion patterns) and the direction of delayed motion during the smoke's ascent, thereby achieving a refined characterization of fire smoke diffusion behavior. In the initial stage of fire smoke, a significant concentration gradient is observed: the central region forms a high-grayscale core area due to high particle density, while the surrounding area shows a decreasing distribution as diffusion progresses. As combustion continues, the smoke area undergoes spatiotemporal evolution: the core area continues to expand, the grayscale value of the original core area significantly increases, forming a new high-concentration center. The overall diffusion area maintains a "center-edge" decreasing pattern, but the coverage area expands compared to the initial stage, and the global maximum grayscale value increases, exhibiting a "core brightening-edge expansion" fire smoke diffusion characteristic. The movement trajectory of the smoke follows the coordinated regulation of environmental airflow and thermodynamic mechanisms. In the initial stage of the smoke movement, it exhibits vertical lifting motion driven by buoyancy. Even if it encounters obstacles that cause local turbulence, its macroscopic motion vector still maintains consistency with the direction of the mainstream field.

[0037] Furthermore, based on the dynamic diffusion characteristics of fire smoke, the determination of real fire smoke areas is achieved through gradient evolution analysis of multiple target images. First, gradient calculation is performed on the smoke areas in each frame of the target image to identify the core region with high concentration. The core region is the connected region with the largest gradient amplitude, while the gradient amplitude of the edge regions decreases with distance. Dynamic analysis is then performed on the core regions across multiple consecutive frames; the area and grayscale of the core regions should increase over time. For example,... Figure 3 As shown, Figure 3 This is a schematic diagram illustrating the changing trend of a core region, provided as an embodiment of the present invention. Figure 3 In the data, the smoke concentration and area of ​​the core area at time t+1 are higher than the smoke concentration of the core area at time t, indicating that the fire is gradually expanding at this time.

[0038] Further, as an optional embodiment of the present invention, determining whether a smoke region in each frame of the target image is a fire smoke region based on the grayscale value of the smoke region in each frame of the target image includes: segmenting the smoke region from each target image based on the grayscale difference between each frame of the target image and the background image; determining the area change trend based on the area of ​​the core region of the smoke region in the target images of adjacent frames, and determining the grayscale change trend based on the average grayscale of the core region of the smoke region in the target images of adjacent frames, wherein the core region is a high grayscale core region where smoke particles are concentrated in the smoke region; determining the smoke region as a suspected fire smoke region if both the area change trend and the grayscale change trend are greater than a second threshold; in the case of a suspected fire smoke region, selecting feature pixels with rich texture from the suspected fire smoke region; determining the angle between the motion vector of the feature pixel and the unit vector in the vertical direction based on the horizontal and vertical components of the feature pixel, wherein the motion vector is determined by comparing the positional changes of the same feature pixel in the target images of adjacent frames; and determining the suspected fire smoke region as a fire smoke region if the proportion of feature pixels with an angle less than a third threshold is greater than or equal to a fourth threshold.

[0039] Specifically, the embodiments of the present invention use the following formula to calculate the area change trend:

[0040]

[0041] In the above formula, This represents the trend of area change in the core region from time t to time t+1. A t This represents the area of ​​the core region in frame t. This represents the area of ​​the core region in frame t+1.

[0042] Furthermore, in this embodiment of the invention, the following formula is used to calculate the grayscale change trend:

[0043]

[0044] In the above formula, This represents the grayscale change trend of the core region from time t to time t+1. μt represents the average gray level of the core region in frame t+1.

[0045] Furthermore, in this embodiment of the invention, the second threshold can be set to 0, and the detection of smoke in the core region satisfies... and If so, the smoke area is marked as a suspected fire smoke area.

[0046] Furthermore, in this embodiment of the invention, key points are extracted within the suspected smoke region of each frame of the target image. A corner detection algorithm is used to select N pixels with rich texture as tracking targets, i.e., feature pixels (such as the corners of the smoke edge). The Lucas-Kanade optical flow algorithm is used to track the displacement of the feature points (feature pixels are simply referred to as feature points) frame by frame. By comparing the positional changes of the same feature point in two adjacent frames of the target image, its horizontal and vertical motion vectors are calculated, and the motion vector of the feature point is synthesized. In this process, feature points that cannot be matched due to occlusion or motion blur are removed, and only continuously successfully tracked points are retained for subsequent analysis. By calculating the motion vectors of N feature points and combining the angle between the motion vectors and the vertical direction vector, it is determined whether the motion direction of the feature points conforms to the motion characteristics of fire smoke. The motion vector of feature point i is... The unit vector in the vertical direction can be represented as The angle θ between the motion vector and the unit vector can be derived from the vector dot product formula.

[0047] Furthermore, as an optional embodiment of the present invention, determining the angle between the motion vector of the feature pixel and the unit vector in the vertical direction based on the horizontal and vertical components of the feature pixel includes: calculating the square root of the sum of the squares of the horizontal and vertical components, and calculating a third ratio between the vertical component and the square root of the sum; processing the third ratio using the inverse cosine function to obtain the angle.

[0048] Specifically, in this embodiment of the invention, the dot product of the motion vector and the unit vector is defined as: .

[0049] Substituting the motion vector into the above formula and vertical unit vector :

[0050] Calculate the dot product:

[0051] Calculate the modulus:

[0052] Furthermore, substituting the dot product and modulus into the above formula:

[0053]

[0054] get .

[0055] In the above formula, x represents the displacement of feature point i in the horizontal direction, i.e., the horizontal component, and y represents the displacement of feature point i in the vertical direction, i.e., the vertical component.

[0056] Furthermore, fire smoke in a warehouse is primarily driven by buoyancy, rising vertically with a high degree of uniformity. Therefore, compared to non-fire smoke, such as dust and water vapor, their movement is more random and directional, with a lower proportion conforming to a vertical direction.

[0057] Furthermore, in this embodiment of the invention, the third threshold can be 20°, and the fourth threshold can be 70%. When the angle θ between the motion direction of feature points and the vertical direction in two consecutive frames of target images within a suspected smoke area is less than 20°, and the proportion of feature points with this angle is ≥70%, it is identified as a fire smoke area.

[0058] Step S102: In the case of a fire smoke area, determine the actual brightness of each sub-area in the fire smoke area based on the original brightness of the light source center in the fire smoke area when there is no smoke, the brightness attenuation coefficient, and the distance.

[0059] Wherein, distance refers to the distance between the position of each sampling point set along the optical axis of the light source center and the center of the light source, and sub-region is formed by dividing the fire smoke area by each sampling point set along the optical axis of the light source center. For example, Figure 4 As shown, Figure 4 This is a schematic diagram illustrating the distribution of sub-regions within a fire smoke area, provided in an embodiment of the present invention. Figure 4 In this embodiment, sampling points are uniformly arranged perpendicular to the central axis, with the maximum horizontal value of the fire smoke area as the boundary. Then, each sampling point is connected to a light source. The areas formed by the lines connecting all sampling points to the light source, as well as the areas between any two adjacent lines on the central axis, within the fire smoke area are called sub-regions. It is worth noting that sampling points can also be set at other locations; this embodiment of the invention does not limit this to specific locations.

[0060] Furthermore, in fixed-light-source scenarios such as warehouses, the quantification of smoke concentration can be achieved by analyzing the natural attenuation law of the light source. The specific method is as follows: First, using a smoke-free background image, the distribution characteristics of the light source are determined—the brightness is highest at the center (optical axis) of the light source, decreasing in a cone shape outwards (darker further away from the center). Based on the segmented smoke area, equidistant sampling is performed along the central axis of the light source in both directions, recording the grayscale value of each sampling point. Since smoke blocks light (the denser the smoke, the less light penetrates), the light intensity is linearly inversely proportional to the propagation distance; along the central axis of the light source, the brightness decreases linearly with the horizontal distance d. The smoke concentration at each sampling point can be deduced; the lower the measured brightness after light penetrates the smoke, the denser the smoke at that location.

[0061] Furthermore, in this embodiment of the invention, firstly, along the central axis of the light source, the brightness decreases linearly with the horizontal distance d. The brightness attenuation coefficient can be determined by linear fitting. As an optional embodiment of the invention, the actual brightness of each sub-region in the fire smoke area is determined based on the original brightness of the center of the light source in the smoke-free area, the brightness attenuation coefficient, and the distance. This includes: calculating the first product between the brightness attenuation coefficient and the absolute value of the distance; and determining the third difference between the original brightness and the first product as the actual brightness.

[0062] Specifically, the embodiments of the present invention use the following formula to calculate the actual brightness:

[0063]

[0064] In the above formula, This represents the actual brightness of the sub-region that is divided when the light travels to a distance d. This represents the original brightness of the light source center when there is no smoke. k represents the brightness attenuation coefficient of the light source. In a warehouse scene where the light source is fixed, k usually takes a value between 1 and 3. , representing the distance between the position of each sampling point set along the optical axis of the light source and the center of the light source. Wherein, Indicates the horizontal coordinates of the sampling point. The horizontal coordinates of the center of the light source.

[0065] Step S103: Determine the global concentration of the fire smoke region based on the original brightness, the actual brightness of each sub-region, and the number of pixels, and determine the instantaneous increase rate of smoke concentration based on the global concentration of the fire smoke region in the target image of the current frame and the global concentration of the fire smoke region in the target image of the previous frame.

[0066] Specifically, the sampling points at each location divide the fire smoke area into i sub-regions along the central axis. The average brightness of each sub-region is calculated to obtain the average brightness of all sub-regions. Smoke concentration C in each sub-regioni It is proportional to the brightness attenuation coefficient.

[0067] Furthermore, as an optional embodiment of the present invention, determining the smoke concentration of each sub-region based on the actual brightness, calibration constant, and original brightness of each sub-region in the fire smoke area includes: calculating the average brightness of the actual brightness of each sub-region in the fire smoke area, and calculating a fifth ratio between the average brightness and the original brightness; calculating a fourth difference between a predetermined value and the fifth ratio; and determining the third product between the calibration constant and the fourth difference as the smoke concentration.

[0068] Specifically, in this embodiment of the invention, the predetermined value is 1, and the smoke concentration is calculated using the following formula:

[0069]

[0070] In the above formula, Let represent the smoke concentration in the i-th sub-region. K represents the calibration constant, which is obtained by the ratio of the measured brightness in the smoke-free state to the theoretical brightness of the attenuation model. This represents the actual brightness of the i-th sub-region, i.e., the original brightness. This represents the average brightness of all sub-regions.

[0071] For example, such as Figure 5 As shown, Figure 5 This is a schematic diagram of an attenuation model provided in an embodiment of the present invention. In this diagram, the light intensity is highest along the central axis and decreases in the horizontal direction of the central axis. In each sub-region, the theoretical brightness of each sub-region is determined according to this decreasing rule.

[0072] Furthermore, in this embodiment of the invention, the global concentration of smoke in the fire smoke area is calculated using the following formula:

[0073]

[0074] In the above formula, This indicates the global concentration of smoke in the fire's smoke area. This indicates the number of sub-regions within the fire smoke area. This represents the number of pixels in the i-th sub-region. This represents the smoke concentration in the i-th sub-region.

[0075] Step S104: Determine the fire risk level by using the instantaneous increase in smoke concentration and trigger the corresponding alarm.

[0076] Specifically, in this embodiment of the invention, after obtaining the quantitative indicators of fire smoke, the smoke concentration of each frame in the multi-frame target image is quantitatively analyzed, and the warehouse fire risk level is classified based on the instantaneous increase in fire smoke concentration between the multi-frame target images, so as to trigger manual intervention and reminders in a timely manner.

[0077] Furthermore, as an optional embodiment of the present invention, determining the instantaneous growth rate of smoke concentration based on the global concentration of the fire smoke region in the target image of the current frame and the global concentration of the fire smoke region in the target image of the previous frame includes: calculating a fifth difference between the global concentration of the fire smoke region in the target image of the current frame and the global concentration of the fire smoke region in the target image of the previous frame, and the time interval between the target image of the current frame and the target image of the previous frame; and determining a sixth ratio between the fifth difference and the time interval as the instantaneous growth rate.

[0078] Specifically, in this embodiment of the invention, the smoke concentration of the target image in the current t-th frame is first obtained. ,like A value of 0 indicates that there is no fire smoke in the target area, meaning there is no fire. When the value is not 0, calculate the instantaneous increase in smoke concentration. The calculation formula is as follows:

[0079]

[0080] In the above formula, This represents the instantaneous increase in smoke concentration in the target image at frame t. This represents the smoke concentration of the target image in frame t. This represents the smoke concentration of the target image in frame t-1.

[0081] Furthermore, as an optional embodiment of the present invention, determining the fire risk level and triggering the corresponding alarm based on the instantaneous increase in smoke concentration includes: when the instantaneous increase in smoke concentration is greater than a fifth threshold and less than a sixth threshold, determining the fire risk level as primary and triggering a primary alarm, and notifying personnel to inspect when a primary alarm occurs; when the instantaneous increase in smoke concentration is greater than or equal to the sixth threshold and less than a seventh threshold, determining the fire risk level as intermediate and triggering a intermediate alarm, and shutting down the ventilation system when a intermediate alarm occurs; when the instantaneous increase in smoke concentration is greater than or equal to the seventh threshold, determining the fire risk level as high and triggering a high alarm, and activating sprinklers when a high alarm occurs.

[0082] Specifically, the fifth, sixth, and seventh thresholds in this embodiment of the invention can be determined according to actual circumstances, and this embodiment of the invention does not impose any limitations on them. The fifth threshold can be 0.2β. critThe sixth threshold can be β crit The seventh threshold can be 2β crit If β t >0.2β crit This triggers a primary alarm, including an audible and visual alarm, notifying on-duty personnel to check the camera footage. crit This represents the critical rate of increase in smoke concentration, specifically determined by the material being burned (natural wood 0.8-1.5, textiles 1.2-1.8, foam 1.8-2.5). If β... t >β crit If this occurs, a medium-level alarm will be triggered, the ventilation system will be shut down, and data will be uploaded to the fire protection platform. t >2β crit If this occurs, an advanced alarm will be triggered, the sprinkler system will be activated, and the fire department will be notified.

[0083] This invention provides a non-contact method for detecting fire smoke particles. It captures images of a target area and determines the fire smoke situation based on differences in grayscale values ​​and brightness within the captured images. By combining multiple captured images, it constructs a fire risk level for the fire smoke and triggers a corresponding alarm. Therefore, compared to using a single sensor, this method is unaffected by complex environments, improving the accuracy and reliability of fire smoke particle detection.

Claims

1. A method for detecting smoke particles in intelligent fire protection, characterized in that, The intelligent fire protection method for detecting smoke particles includes: Acquire multiple frames of target images of the target area, and determine whether the smoke area in each frame of the target image is a fire smoke area based on the gray value of the smoke area in each frame of the target image. In the case of the fire smoke area, the actual brightness of each sub-region in the fire smoke area is determined based on the original brightness, brightness attenuation coefficient and distance of the light source center in the fire smoke area when there is no smoke. The distance is the distance between the position of each sampling point set along the optical axis of the light source center and the light source center. The sub-region is formed by dividing the fire smoke area by each sampling point set along the optical axis of the light source center. The global concentration of the fire smoke region is determined based on the original brightness, the actual brightness of each sub-region, and the number of pixels. The instantaneous rate of increase of the smoke concentration is determined based on the global concentration of the fire smoke region in the target image of the current frame and the global concentration of the fire smoke region in the target image of the previous frame. The fire risk level is determined by the instantaneous increase in smoke concentration, and the corresponding alarm level is triggered. Determining the global concentration of the fire smoke area based on the original brightness, the actual brightness of each sub-region, and the number of pixels includes: The smoke concentration of each sub-region in the fire smoke region is determined based on the actual brightness, calibration constant, and original brightness of each sub-region. Calculate the second product between the smoke concentration of each sub-region and the corresponding number of pixels, and then sum the second products to obtain a first summed value; The pixel counts of each sub-region are superimposed to obtain a second superimposed value; The fourth ratio between the first superimposed value and the second superimposed value is determined as the global concentration of the fire smoke area; The step of determining the smoke concentration of each sub-region based on the actual brightness, calibration constant, and original brightness of each sub-region in the fire smoke region includes: Calculate the average brightness of the actual brightness of each sub-region in the fire smoke area, and calculate the fifth ratio between the average brightness and the original brightness; Calculate the fourth difference between the predetermined value and the fifth ratio, where the predetermined value is 1; The third product between the calibration constant and the fourth difference is determined as the smoke concentration.

2. The fire smoke particle detection method for intelligent fire protection according to claim 1, characterized in that, The acquisition of multiple frames of target images of the target region includes: The image of the target area captured by the camera is compared with the background image, and the pixel-level difference between the current frame image and the background image is calculated by the image difference method. If the pixel-level difference exceeds the first threshold, it is determined that the target area is suspected of being a fire. The target area is captured by the camera at high frame rate to obtain the target image. The frame rate of the high frame rate capture is the ratio between the smoke diffusion speed and the maximum allowable displacement per frame.

3. The fire smoke particle detection method for intelligent fire protection according to claim 1, characterized in that, The step of determining whether a smoke region in each frame of the target image is a fire smoke region based on the grayscale value of the smoke region in each frame of the target image includes: Smoke regions are segmented from each target image based on the grayscale differences between the target image and the background image in each frame; The area change trend is determined based on the area of ​​the core region of the smoke region in the target image of adjacent frames, and the gray value change trend is determined based on the average gray value of the core region of the smoke region in the target image of adjacent frames. The core region is the high gray value core region where smoke particles are concentrated in the smoke region. If the area change trend is greater than the second threshold and the grayscale change trend is greater than the second threshold, the smoke area is determined to be a suspected fire smoke area. In the case of the suspected fire smoke area, feature pixels with rich texture are selected from the suspected fire smoke area; The angle between the motion vector of the feature pixel and the unit vector in the vertical direction is determined based on the horizontal and vertical components of the feature pixel. The motion vector is determined by comparing the positional changes of the same feature pixel in the target image of adjacent frames. If the proportion of feature pixels with an angle less than the third threshold is greater than or equal to the fourth threshold, the suspected fire smoke area is determined to be a fire smoke area.

4. The fire smoke particle detection method for intelligent fire protection according to claim 3, characterized in that, The method of determining the area change trend based on the area of ​​the core region of the smoke region in the target image of adjacent frames includes: Calculate the first difference between the area of ​​the core region of the smoke region in the target image of the next frame and the area of ​​the core region of the smoke region in the target image of the current frame; The first ratio between the first difference and the area of ​​the core region of the smoke region in the target image of the current frame is determined as the trend of the area change of the core region of the smoke region in the target image of the current frame. The step of determining the grayscale change trend based on the average grayscale of the core region of the smoke region in the target image of adjacent frames includes: Calculate the second difference between the average grayscale of the core region of the smoke region in the target image of the next frame and the average grayscale of the core region of the smoke region in the target image of the current frame. The second ratio between the second difference and the average gray level of the core region of the smoke region in the target image of the current frame is determined as the gray level change trend of the core region of the smoke region in the target image of the current frame.

5. The fire smoke particle detection method for intelligent fire protection according to claim 3, characterized in that, The step of determining the angle between the motion vector of the feature pixel and the unit vector in the vertical direction based on the horizontal and vertical components of the feature pixel includes: Calculate the square root of the sum of the squares of the horizontal component and the squares of the vertical component, and calculate the third ratio between the vertical component and the square root of the sum; The included angle is obtained by processing the third ratio using the inverse cosine function.

6. The fire smoke particle detection method for intelligent fire protection according to claim 1, characterized in that, The step of determining the actual brightness of each sub-region within the fire smoke region based on the original brightness, brightness attenuation coefficient, and distance of the light source center in the smoke-free area includes: Calculate the first product between the brightness attenuation coefficient and the absolute value of the distance; The third difference between the original brightness and the first product is determined as the actual brightness.

7. The intelligent fire protection smoke particle detection method according to claim 1, characterized in that, The instantaneous rate of increase in smoke concentration, determined based on the global concentration of the fire smoke region in the target image of the current frame and the global concentration of the fire smoke region in the target image of the previous frame, includes: Calculate the fifth difference between the global concentration of the fire smoke region in the target image of the current frame and the global concentration of the fire smoke region in the target image of the previous frame, as well as the time interval between the target image of the current frame and the target image of the previous frame; The sixth ratio between the fifth difference and the time interval is determined as the instantaneous growth rate.

8. The fire smoke particle detection method for intelligent fire protection according to claim 1, characterized in that, The method of determining the fire risk level and triggering the corresponding alarm based on the instantaneous increase in smoke concentration includes: If the instantaneous increase in smoke concentration is greater than the fifth threshold but less than the sixth threshold, the fire risk level is determined to be primary and a primary alarm is triggered. When a primary alarm is triggered, staff are notified to check. If the instantaneous increase in smoke concentration is greater than or equal to the sixth threshold and less than the seventh threshold, the fire risk level is determined to be medium and a medium alarm is triggered. When a medium alarm is triggered, the ventilation system is shut down. If the instantaneous increase in smoke concentration is greater than or equal to the seventh threshold, the fire risk level is determined to be high and a high-level alarm is triggered. When a high-level alarm is triggered, the sprinkler system is activated.

Images