Apparatus and method for overhead power line passage pyrotechnic hazard identification

Abstract

The device for identifying smoke and fire hidden dangers in an overhead power transmission line passage comprises a storage unit configured to store a standard image associated with a target scene, the standard image having at least a first standard region and a second standard region, a processing priority of the first standard region being higher than a processing priority of the second standard region; and a processing unit configured to: acquire a current video frame associated with the target scene; acquire a first detection region associated with the first standard region in the current video frame; acquire a first intersection between the first detection region and the first standard region; calculate a first ratio between the first intersection and the first detection region; if the first ratio is greater than or equal to a first effective threshold, maintain or increase an alarm level of the first detection region; otherwise, decrease the alarm level of the first detection region. A method is also provided. The present application can reduce the adverse effects of false alarms and improve the efficiency of power transmission line operation and inspection.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of intelligent detection, and particularly relates to a device for identifying smoke and fire hazards in an overhead power transmission line passage, and a method for identifying smoke and fire hazards in an overhead power transmission line passage. BACKGROUND

[0002] Power system transmission operation and inspection is a guarantee for efficient and safe operation of power transmission. In the field of intelligent transmission operation and inspection, the identification accuracy of hidden hazards has reached a relatively ideal level, and hidden hazards that may cause harm to the power transmission line, such as large mechanical equipment and fireworks, can be accurately identified. Among various hidden hazards, fireworks in dry climates in fields, mountains and forests are highly dangerous to power transmission lines and are the focus of attention in power transmission line hidden hazard detection.

[0003] The prior art discloses various methods to achieve smoke detection, such as the scheme disclosed in Chinese Patent Application (CN1098A5863A): “A smoke detection method based on deep learning and image recognition, comprising the following steps: Step 1, reading video frames from the video to be identified, extracting part of the pixel set in the video frame according to the color features of the pixels in the video frame and the difference features of the pixels between video frames, forming one or more pre-processing areas for identification; Step 2, cutting the extracted pre-processing area into a picture as the input of the smoke detection model; Step 3, extracting features from the picture through the smoke detection model; Step 4, determining the confidence of the presence of a smoke scene in the picture according to the extracted features; Step 5, if the confidence exceeds the confidence threshold, marking the area corresponding to the picture in the video frame.” … and further discloses: “Step 3a, generating a feature extraction network to extract image features; Step 3b, collecting pictures containing smoke, labeling them to form a data set, and dividing them into a training set and a test set according to a certain proportion; Step 3c, using the training set as the input of the smoke detection model to train the model; Step 3d, testing the effect of the smoke detection model using the test set, and adjusting the model parameters at the same time.”

[0004] The research idea of the prior art is to learn a large amount of data to improve the performance of the model, and let the model find the differences between smoke and other objects, so as to reduce the false alarm rate of smoke hazards. However, this method requires a large amount of data to support, and is obviously affected by the environment. For example, in some scenarios, the features of clouds in the sky and smoke, sun halos, bright lights and fire features in night scenes are very similar, and the probability of false alarms is high.

[0005] The above information disclosed in the background of the disclosure is only used to increase the understanding of the background of the disclosure, and therefore, it can include prior art known to those of ordinary skill in the art. SUMMARY

[0006] The present application is directed to the prior art by finding the difference between fireworks and other object features through the model, so as to reduce the false alarm rate of fireworks hazards, but such method needs a large amount of data to support, and is obviously affected by the environment, especially clouds, halos, lights and the like, and the probability of false alarm is relatively high, and one aspect designs and provides a device for identifying smoke and fire hazards in overhead power transmission line channel.

[0007] The device for identifying smoke and fire hazards in overhead power transmission line channel comprises a storage unit configured to store a standard image associated with at least one target scene, the standard image having at least a first standard region and a second standard region, the processing priority of the first standard region being higher than that of the second standard region; and a processing unit configured to: acquire a current video frame associated with the target scene; acquire a first detection region associated with the first standard region in the current video frame; acquire a first intersection between the first standard region and the first detection region; calculate a first ratio between the first intersection and the first detection region; if the first ratio is greater than or equal to a first effective threshold, maintain or improve the alarm level of the first detection region; if the first ratio is less than the first effective threshold, reduce the alarm level of the first detection region.

[0008] The second aspect of the present application provides a method for identifying smoke and fire hazards in overhead power transmission line channel, specifically comprising the following steps: acquiring a current video frame of at least one target scene; calling a standard image associated with the target scene, the standard image having at least a first standard region and a second standard region, the processing priority of the first standard region being higher than that of the second standard region; acquiring a first detection region associated with the first standard region in the current video frame; acquiring a first intersection between the first standard region and the first detection region; calculating a first ratio between the first intersection and the first detection region; if the first ratio is greater than or equal to a first effective threshold, maintaining or improving the alarm level of the first detection region; if the first ratio is less than the first effective threshold, reducing the alarm level of the first detection region.

[0009] Compared with the prior art, the present application has the following advantages and positive effects:

[0010] The device and method for identifying smoke and fire hazards in overhead power transmission line channel provided by the present application can accurately calibrate the processing priority of the standard region in the target scene and calibrate according to consecutive video frames, so that the system image processing model can concentrate computing power on the region with higher processing priority, thereby improving the processing speed and efficiency of the system. At the same time, the interference of lights, halos and clouds can be accurately filtered out, the adverse effects caused by false alarms are reduced, and the inspection efficiency of the power transmission line is improved.

[0011] Other features and advantages of the present application will become more apparent from the following detailed description when read in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort based on these drawings.

[0013] Figure 1 A structural schematic block diagram of the device for identifying smoke and fire hazards in an overhead power transmission line passage provided by the present application is shown in FIG. 1.

[0014] Figure 2 A flowchart of the method for identifying smoke and fire hazards in an overhead power transmission line passage provided by the present application is shown in FIG. 2.

[0015] Figure 3 A flowchart of the method for identifying smoke and fire hazards in an overhead power transmission line passage provided by the present application after comparing the first ratio with the first effective threshold value is shown in FIG. 3.

[0016] Figure 4 A flowchart of the method for identifying smoke and fire hazards in an overhead power transmission line passage provided by the present application after assigning the same hazard object identifier to the first detection area and the second detection area is shown in FIG. 4. DETAILED DESCRIPTION

[0017] In order to make the objectives, technical solutions and advantages of the present application more clear, the present application will be further described in detail below with reference to the drawings and embodiments.

[0018] The terms "first", "second", "third", etc. in the specification and claims of the present application and the drawings are used to distinguish different objects, and are not used to describe a specific order. In addition, the terms "include" and "have" and any variations thereof represent an inclusive rather than an exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but can optionally include other steps or units not listed or can optionally include other steps or units inherent to the process, method, product or device.

[0019] In this disclosure "an embodiment" or "one embodiment" or "an implementation" or "one implementation" or the like means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. The appearances of the phrase "in one embodiment" or "in an embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are they necessarily all directed to the same embodiment, nor are they necessarily all mutually exclusive embodiments. One skilled in the art will recognize that the embodiments described herein can be combined with other embodiments in various ways.

[0020] Figure 1 A schematic block diagram of a principle of an apparatus for overhead transmission line passage pyrotechnic hazard recognition according to one or more embodiments described herein is depicted. The apparatus is intended to run independently or work as an auxiliary or redundant recognition apparatus of a model to solve the false alarm rate caused by environmental factors (clouds, halos, lights, etc.). From the main point of view, the apparatus is composed of a storage unit and a processing unit. The storage unit is configured to store a standard image associated with at least one target scene, the standard image having at least a first standard region and a second standard region, wherein the processing priority of the first standard region is higher than that of the second standard region. The standard image is an image of the outside and the vicinity of the overhead transmission line passage, which is collected by a camera. The number of cameras can be one or more, which is not limited here. The standard image is preferably selected from the image under sunny and non-foggy weather conditions, and there is no any pyrotechnic hazard in the standard image, and the processing priority is the lowest. The standard image has at least a first standard region and a second standard region, the first standard region corresponds to a region with dense pyrotechnic hazards, such as mountains and fields. The second standard region corresponds to a region with sparse pyrotechnic hazards, such as the sky and the lake, and the processing priority of the first standard region is higher than that of the second standard region. The standard image can also be divided into more levels of standard regions, for example, the first standard region corresponds to mountains, the second standard region corresponds to fields, the third standard region corresponds to the sky, and the fourth standard region corresponds to lakes. The division of the standard region can be completed by artificial or by a graphic processing software according to color or texture features. The first standard region and the second standard region are selected in the form of a rectangular frame, and the number of the first standard region and the second standard region is at least one.

[0021] As Figure 1The second main component of the device is the processing unit. The processing unit is configured to work according to the following flow when identifying a possible fire hazard in the target scene, where the identification of the fire hazard can be achieved by manual or by existing image processing models. The working process of the processing unit specifically includes: obtaining a current video frame associated with the target scene. The current video frame and the standard image are captured by the same camera. Obtain a first detection region associated with the first standard region in the current video frame. The first detection region is obtained in the form of a rectangular box, and the aspect ratio and area of the first detection region are the same as those of the first standard region. The first detection region frames the fire hazard. Obtain the first intersection between the first standard region and the first detection region, and calculate the first ratio between the first intersection and the first detection region. The processing unit is further configured to determine whether the first ratio is greater than or equal to the first effective threshold value. If the first ratio is greater than the first effective threshold value, it is determined that the fire hazard is in the first detection region, and the warning level of the first detection region is maintained or increased. If the first ratio is less than the first effective threshold value, it is determined that the fire hazard deviates from the first detection region. After the determination is completed, the processing unit can automatically obtain the detection region with the highest warning level and automatically filter out the detection region with the lowest warning level, thereby concentrating resources on processing regions with higher warning levels and improving the processing accuracy and speed of the system.

[0022] To reduce the misjudgment rate, the processing unit is further configured to: obtain a first current video frame and a second current video frame associated with the target scene. The first current video frame and the second current video frame are obtained sequentially along the time dimension, and the first current video frame is earlier than the second current video frame. The time interval between the first current video frame and the second current video frame is less than or equal to 1 minute. Obtain a first detection region associated with the first standard region in the first current video frame, and obtain a second detection region associated with the second standard region in the second current video frame. Obtain a second intersection between the first detection region and the second detection region. Calculate the second ratio between the second intersection and the second detection region. If the second ratio is greater than or equal to the second effective threshold value, assign the same hazard object identifier to the first detection region and the second detection region, i.e. there is the same fire hazard in the first detection region and the second detection region. If the second ratio is less than the second effective threshold value, the first current video frame and the second current video frame are obtained sequentially along the time dimension again.

[0023] Since the first effective threshold value is used to determine the processing priority, and the second effective threshold value is used to determine whether there is the same hazard point in the detection region with the highest processing priority in the two consecutive video frames, the determination accuracy of the latter needs to be higher than that of the former, so the first effective threshold value is set to be less than the second effective threshold value, for example, the first effective threshold value is set to 0.6 and the second effective threshold value is set to 0.8.

[0024] To eliminate the influence of clouds, halos or lights, after assigning the same hazard object identifier to the first detection area and the second detection area, the processing unit is further configured to: obtain a first landmark point coordinate in the first detection area, and obtain a second landmark point coordinate associated with the first landmark point in the second detection area. If the offset of the first landmark point and the second landmark point in any one of the length direction or the width direction is less than the first offset threshold, it indicates that the fireworks hazard corresponding to the first landmark point has not moved in a certain time interval, or the moving distance does not match the dynamic characteristics of the fireworks hazard, thereby further reducing the warning level of the first detection area. If the offset of the first landmark point and the second landmark point in any one of the length direction or the width direction is greater than the first offset threshold, and the dynamic movement characteristics of the fireworks hazard corresponding to the first landmark point match the dynamic movement characteristics of the fireworks hazard, the warning level of the first detection area is maintained or improved.

[0025] The first offset threshold is preferably set according to the sampling time interval between the first current video frame and the second current video frame. The longer the sampling time interval, the larger the first offset threshold, i.e., the first offset threshold is positively correlated with the sampling time interval. The first offset threshold is preferably set as a dynamic pixel value, specifically the product of a reference pixel and a current wind speed adjustment parameter. The current wind speed adjustment parameter is generated according to the wind speed of the first current video frame and the second current video frame. The larger the wind speed, the larger the wind speed adjustment parameter. The increase amplitude of the wind speed adjustment parameter is greater than that of the wind speed, so as to filter out the interference of clouds that can also move with the wind. The reference pixel is preferably set to 5 pixel values. The wind speed is preferably measured by a wind speed meter arranged in the target scene. The first landmark point can be a pixel point corresponding to the fireworks hazard target, and is preferably set as the center point of the first detection area. The second landmark point is preferably the center point of the second detection area. The length-width ratio of the first detection area and the second detection area is between 0.9-1.1.

[0026] The device for identifying fireworks hazards in an overhead power transmission line passage provided by the present application can accurately calibrate the processing priority of the standard area in the target scene and calibrate according to consecutive video frames, so that the system image processing model can concentrate computing power on areas with higher processing priority, thereby improving the processing speed and efficiency of the system. At the same time, it can accurately filter out the interference of lights, halos and clouds, reduce the adverse effects of false alarms, and improve the inspection efficiency of the power transmission line.

[0027] Figure 2 A flowchart of an overhead power transmission line passage fireworks hazard identification method according to one or more embodiments described herein is depicted. The method aims to improve the accuracy of fireworks hazard identification. The overhead power transmission line passage fireworks hazard identification method specifically includes the following steps as detailed below.

[0028] Step S11: Obtain a current video frame of at least one target scene. The target scene is an area outside and near the overhead transmission line channel, and the current video frame is obtained based on dynamic video or static pictures collected by a camera. The number of cameras can be one or more, which is not limited herein.

[0029] Step S12: Call a standard image associated with the target scene. The standard image has at least a first standard region and a second standard region, and the processing priority of the first standard region is higher than that of the second standard region. The standard image is an image obtained in a sunny and non-foggy weather condition, and there is no any fire hazard in the standard image, and the processing priority is the lowest. The first standard region corresponds to an area where the fire hazard is more likely to appear and is dense, such as mountains and antennas, and the second standard region corresponds to an area where the fire hazard is less likely to appear and is sparse, such as the sky and lakes. More levels of standard regions can be divided in the standard image, such as the first standard region corresponding to mountains, the second standard region corresponding to fields, the third standard region corresponding to the sky, and the fourth standard region corresponding to lakes. The division of the standard region can be completed by artificial or by a graphic processing software according to color or texture features. The first standard region and the second standard region are selected in the form of a rectangular frame, and the number of the first standard region and the second standard region is at least one.

[0030] Step S13: Obtain a first detection region associated with the first standard region in the current video frame. The first detection region is selected in the form of a rectangular frame, the aspect ratio and the area of the first detection region are the same as those of the first standard region, and the fire hazard is framed in the first detection region. The characteristics of the fire hazard itself refer to the determination conditions of the classifier in the prior art.

[0031] Step S14: Obtain a first intersection between the first standard region and the first detection region.

[0032] Step S15: Calculate a first ratio between the first intersection and the first detection region.

[0033] Step S16: If the first ratio is greater than or equal to a first effective threshold, it is determined that the fire hazard is in the first detection region, and the warning level of the first detection region is maintained or improved.

[0034] Step S17: If the first ratio is less than the first effective threshold, it is determined that the fire hazard deviates from the first detection region, and the warning level of the first detection region is reduced. After the determination is completed, the detection region with the highest warning level can be automatically obtained, and the detection region with the lowest warning level can be automatically filtered out, so as to concentrate resources and computing power on processing the regions with higher warning levels, and improve the processing accuracy and processing speed of the system.

[0035] In order to reduce the misjudgment rate,Figure 3 As shown, after comparing the first ratio with the first effective threshold, the overhead transmission line passage fireworks hidden danger identification method further comprises the following steps:

[0036] Step S21: Obtain a first current video frame and a second current video frame associated with a target scene; the first current video frame and the second current video frame are obtained in sequence along a time dimension, and the first current video frame is earlier than the second current video frame. The time interval between the first current video frame and the second current video frame is less than or equal to 1 minute.

[0037] Step S22: Obtain a first detection region associated with a first standard region in the first current video frame. The first detection region is preferably framed with a rectangular box, and the aspect ratio and area of the first detection region are preferably the same as those of the first standard region. The first detection region frames the fireworks hidden danger.

[0038] Step S23: Obtain a second detection region associated with a second standard region in the second current video frame. The second detection region is also preferably framed with a rectangular box, and the aspect ratio and area of the second detection region are preferably the same as those of the first standard region. The second detection region frames the fireworks hidden danger.

[0039] Step S24: Calculate a second ratio between the second intersection and the second detection region.

[0040] Step S25: If the second ratio is greater than or equal to a second effective threshold, assign the same hidden object identifier to the first detection region and the second detection region, indicating that the same fireworks hidden danger exists in the first detection region and the second detection region. If the second ratio is less than the second effective threshold, obtain the first current video frame and the second current video frame in sequence along the time dimension again.

[0041] Since the first effective threshold is used to determine the processing priority, and the second effective threshold is used to determine whether the same hidden point exists in the detection region with the highest processing priority in the two consecutive frames of video, the judgment accuracy of the latter needs to be higher than that of the former, so it is preferred to set the first effective threshold to be less than the second effective threshold, for example, set the first effective threshold to be 0.6 and the second effective threshold to be 0.8.

[0042] In order to eliminate the interference factors introduced by clouds, halos or lights, etc. Figure 4 As shown, after assigning the same hidden object identifier to the first detection region and the second detection region, the overhead transmission line passage fireworks hidden danger identification method further comprises the following steps:

[0043] Step S31: Obtain a first landmark point coordinate in the first detection region.

[0044] Step S32: Obtain a second landmark point coordinate associated with the first landmark point in the second detection region.

[0045] Step S33: If the offset of the first mark point and the second mark point in any one of the length direction or the width direction is less than the first offset threshold, it indicates that the pyrotechnic hazard corresponding to the first mark point does not move in a certain time interval, or the moving distance does not match the dynamic characteristics of the pyrotechnic hazard, and the warning level of the first detection area is further reduced. If the offset of the first mark point and the second mark point in any one of the length or width direction is greater than or equal to the first offset threshold, the dynamic moving characteristics of the pyrotechnic hazard corresponding to the first mark point are consistent with the dynamic moving characteristics of the pyrotechnic hazard, and the warning level of the first detection area is maintained or improved.

[0046] The first offset threshold is preferably set according to the sampling time interval between the first current video frame and the second current video frame. The longer the sampling time interval is, the greater the first offset threshold is, that is, the first offset threshold is positively correlated with the sampling time interval. The first offset threshold is preferably set as a dynamic pixel value, specifically the product between the reference pixel and the current wind speed adjustment parameter. The current wind speed adjustment parameter is generated according to the wind speed of the first current video frame and the second current video frame, and the greater the wind speed is, the greater the wind speed adjustment parameter is. At the same time, the increment of the wind speed adjustment parameter is greater than the increment of the wind speed, so as to reduce the weight introduced by the cloud that can also move with the wind. The reference pixel is preferably set as 5 pixel values. The wind speed is preferably measured by a wind speed meter arranged in the target scene. The first mark point can be a pixel point in the pyrotechnic hazard target, and is preferably set as the center point of the first detection area. If the first mark point is the center point of the first detection area, the second mark point is also preferably the center point of the second detection area. The length-width ratio of the first detection area and the second detection area is between 0.9-1.1.

[0047] The embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program exchanged by electronic data, and the computer program enables a computer to execute part or all steps of any method described in the above method embodiment.

[0048] In the above embodiments, the description of each embodiment is focused on, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0049] In several embodiments provided in the present application, it should be understood that the disclosed apparatus can be implemented in other manners. For example, the division of the apparatus embodiments described above is merely illustrative, and the division of the units or modules can be changed according to actual needs. For example, two or more units or modules can be combined into one unit or module, or some features can be ignored or not executed. In addition, the displayed or discussed coupling or direct coupling or communication connection between units can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical or other forms.

[0050] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one physical space or distributed on multiple network units, and some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0051] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can be physically present separately, or two or more units can be integrated into one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0052] The above embodiments are merely used to illustrate the technical solutions of the present application, rather than limit the same. Although the present application has been described in detail with reference to the foregoing embodiments, the technical solutions recorded in the foregoing embodiments can still be modified by those skilled in the art, or some technical features can be replaced by equivalent ones. Such modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions claimed by the present application.

Claims

1. Equipment for identifying smoke and fire hazards in overhead transmission line corridors, characterized in that, include: A storage unit is configured to store a standard image associated with at least one target scene, the standard image having at least a first standard region and a second standard region; wherein the first standard region corresponds to an area with dense smoke and fire hazards, and the second standard region corresponds to an area with sparse smoke and fire hazards; the processing priority of the first standard region is higher than the processing priority of the second standard region; and Processing unit, the processing unit is configured as follows: Acquire the current video frame associated with the target scene, wherein the current video frame and the standard image are captured by the same camera; In the current video frame, a first detection region associated with the first standard region is obtained: the first detection region is selected in the form of a rectangular frame and the smoke and fire hazard is selected by the first detection region; the aspect ratio and area of ​​the first detection region are the same as those of the first standard region. Obtain the first intersection between the first standard region and the first detection region; Calculate the first ratio between the first intersection and the first detection region; If the first ratio is greater than or equal to the first effective threshold, it is determined that the fire hazard is within the first detection area, and the warning level of the first detection area is maintained or increased; if the first ratio is less than the first effective threshold, it is determined that the fire hazard deviates from the first detection area, and the warning level of the first detection area is reduced. After the determination is completed, the processing unit obtains the detection area with the highest warning level and filters out the detection area with the lowest warning level.

2. The device according to claim 1, characterized in that, The processing unit is further configured to: Acquire a first current video frame and a second current video frame associated with the target scene. The first current video frame and the second current video frame are acquired sequentially along the time dimension, and the first current video frame is earlier than the second current video frame. In the first current video frame, obtain the first detection region associated with the first standard region; In the second current video frame, a second detection region associated with the second standard region is obtained. The second detection region is selected in the form of a rectangular box selection and the smoke and fire hazard is selected by the second detection region. The aspect ratio and area of ​​the second detection region are the same as those of the first standard region. Obtain the second intersection between the first detection region and the second detection region; Calculate the second ratio between the second intersection and the second detection region; If the second ratio is greater than or equal to the second effective threshold, the first detection area and the second detection area are assigned the same hidden danger object identifier; if the second ratio is less than the second effective threshold, the first current video frame and the second current video frame are obtained again along the time dimension.

3. The device according to claim 2, characterized in that, The first effective threshold is less than the second effective threshold.

4. The device according to claim 2, characterized in that, After assigning the same hazard object identifier to both the first detection area and the second detection area, the processing unit is further configured to: Obtain the coordinates of the first marker point in the first detection area; In the second detection area, obtain the coordinates of a second marker point associated with the first marker point; If the offset of the first marker point and the second marker point in either the length direction or the width direction is less than the first offset threshold, then the warning level of the first detection area is reduced. If the offset of the first marker point and the second marker point in either the length direction or the width direction is greater than or equal to the first offset threshold, then the warning level of the first detection area is maintained or increased.

5. The device according to claim 4, characterized in that, The first marker point is the center point of the first detection area, the second marker point is the center point of the second detection area, and the aspect ratio of the first detection area and the second detection area is between 0.9 and 1.

1.

6. A method for identifying smoke and fire hazards in overhead transmission line corridors, characterized in that, Includes the following steps: Acquire the current video frame of at least one target scene; A standard image associated with the target scene is invoked, the standard image having at least a first standard region and a second standard region; wherein, the first standard region corresponds to an area with dense smoke and fire hazards, and the second standard region corresponds to an area with sparse smoke and fire hazards; the processing priority of the first standard region is higher than that of the second standard region; the current video frame and the standard image are captured by the same camera; In the current video frame, a first detection region associated with the first standard region is obtained. The first detection region is selected in the form of a rectangular frame and the smoke and fire hazard is selected by the first detection region. The aspect ratio and area of ​​the first detection region are the same as those of the first standard region. Obtain the first intersection between the first standard region and the first detection region; Calculate the first ratio between the first intersection and the first detection region; If the first ratio is greater than or equal to the first effective threshold, it is determined that the fire hazard is within the first detection area, and the warning level of the first detection area is maintained or increased. If the first ratio is less than the first effective threshold, it is determined that the fire hazard has deviated from the first detection area, and the warning level of the first detection area is reduced. After the determination is completed, the detection area with the highest warning level is obtained, and the detection area with the lowest warning level is filtered out.

7. The method according to claim 6, characterized in that, After comparing the first ratio with the first effective threshold, the method further includes the following steps: Acquire a first current video frame and a second current video frame associated with the target scene. The first current video frame and the second current video frame are acquired sequentially along the time dimension, and the first current video frame is earlier than the second current video frame. In the first current video frame, obtain the first detection region associated with the first standard region; In the second current video frame, a second detection region associated with the second standard region is obtained. The second detection region is selected in the form of a rectangular box selection and the smoke and fire hazard is selected by the second detection region. The aspect ratio and area of ​​the second detection region are the same as those of the first standard region. Obtain the second intersection between the first detection region and the second detection region; Calculate the second ratio between the second intersection and the second detection region; If the second ratio is greater than or equal to the second effective threshold, the first detection area and the second detection area are assigned the same hidden danger object identifier; if the second ratio is less than the second effective threshold, the first current video frame and the second current video frame are obtained again along the time dimension.

8. The method according to claim 7, characterized in that, The first effective threshold is less than the second effective threshold.

9. The method according to claim 7, characterized in that, After assigning the same hazard object identifier to both the first detection area and the second detection area, the method further includes: Obtain the coordinates of the first marker point in the first detection area; In the second detection area, obtain the coordinates of a second marker point associated with the first marker point; If the offset of the first marker point and the second marker point in either the length or width direction is less than the first offset threshold, the warning level of the first detection area is reduced; if the offset of the first marker point and the second marker point in either the length or width direction is greater than or equal to the first offset threshold, the warning level of the first detection area is maintained or increased.

10. The method according to claim 9, characterized in that, The first marker point is the center point of the first detection area, the second marker point is the center point of the second detection area, and the aspect ratio of the first detection area and the second detection area is between 0.9 and 1.1.

Images