Fire information detection method and device, storage medium and electronic equipment

Abstract

The invention discloses a fire information detection method and device, a storage medium and electronic equipment. Relates to the technical field of artificial intelligence. Each video frame in the target video data is processed through a target detection algorithm to obtain a plurality of video frames, the video frames of the plurality of video frames comprise a rectangular frame added by the target detection algorithm, and the probability that a fire exists at a position corresponding to the rectangular frame in each target video frame is greater than a preset threshold value; according to the plurality of video frames, judging whether a fire exists in the target area or not; and under the condition that the fire exists in the target area, determining the position information of the fire according to the plurality of video frames. According to the invention, the problem of low fire detection accuracy caused by fire detection through a temperature-sensing detector, a smoke-sensing detector and the like in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method and device for detecting fire information, a storage medium, and an electronic device. Background Art

[0002] At present, in order to prevent fire problems in branches or large computer rooms, financial institutions have developed many fire detection technologies. For example, temperature or smoke detectors are used to detect the temperature or smoke of branches or large computer rooms to determine whether there is a fire. However, these detectors are difficult to play their due role in places such as branches or large computer rooms, because these detectors will only sound an alarm when the temperature or smoke concentration reaches a certain level. Therefore, the alarm issued is bound to have a certain delay, which is not conducive to the early detection of fire.

[0003] Currently, no effective solution has been proposed to address the problem that the accuracy of fire detection is relatively low when fire detection is performed using temperature and smoke detectors in related technologies. Summary of the invention

[0004] The main purpose of this application is to provide a fire information detection method and device, storage medium and electronic equipment to solve the problem of low accuracy of fire detection caused by fire detection through temperature, smoke and other detectors in related technologies.

[0005] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a method for detecting fire information is provided. The method comprises: obtaining target video data of a target area; processing each video frame in the target video data by a target detection algorithm to obtain multiple video frames, wherein the video frames of the multiple video frames include rectangular boxes added by the target detection algorithm, and the probability that a fire exists at a position corresponding to the rectangular box in each target video frame is greater than a preset threshold; judging whether there is a fire in the target area based on the multiple video frames; and determining the location information of the fire based on the multiple video frames if there is a fire in the target area.

[0006] Furthermore, judging whether there is a fire in the target area based on the multiple video frames includes: for a first video frame, obtaining a first eigenvalue based on color information of an area corresponding to a rectangular box in the first video frame, wherein the first video frame is any one of the multiple video frames; obtaining a second eigenvalue based on area information of an area corresponding to the rectangular box in the first video frame; calculating a third eigenvalue by calculating a similarity between an area corresponding to the rectangular box in the first video frame and an area corresponding to a rectangular box in a plurality of second video frames, wherein the plurality of second video frames are continuous with the first video frame in a time dimension; obtaining a fourth eigenvalue based on shape information of an area corresponding to the rectangular box in the first video frame; judging whether there is a fire in the target area based on the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue.

[0007] Furthermore, according to the color information of the area corresponding to the rectangular box in the first video frame, obtaining the first eigenvalue includes: calculating the three primary colors corresponding to each pixel point in the first area to obtain a red value, a green value and a blue value, wherein the first area is the area corresponding to the rectangular box in the first video frame; calculating according to the red value, the green value and the blue value to obtain a first red proportion value corresponding to the first area; determining the multiple second video frames, and obtaining the second red proportion value of the area corresponding to the rectangular box in each second video frame; calculating the red proportion growth rate according to the first red proportion value and the second red proportion value to obtain the first eigenvalue.

[0008] Furthermore, based on the area information of the area corresponding to the rectangular box in the first video frame, obtaining the second eigenvalue includes: calculating the area of ​​the area corresponding to the rectangular box in the first video frame to obtain a first area value; determining the multiple second video frames, and obtaining the second area value of the area corresponding to the rectangular box in each second video frame; calculating the area growth rate based on the first area value and the second area value to obtain the second eigenvalue.

[0009] Furthermore, based on the shape information of the area corresponding to the rectangular box in the first video frame, obtaining the fourth eigenvalue includes: calculating the area of ​​the first area to obtain a first area value, wherein the first area is the area corresponding to the rectangular box in the first video frame; calculating the perimeter of the first area to obtain a perimeter value; and calculating based on the first area value and the perimeter value to obtain the fourth eigenvalue.

[0010] Further, judging whether there is a fire in the target area based on the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue includes: judging whether the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue meet a preset condition; if the preset condition is met, there is a fire in the target area; if the preset condition is not met, there is no fire in the target area.

[0011] Furthermore, each video frame in the target video data is processed by a target detection algorithm to obtain multiple video frames, including: performing feature extraction on each video frame by a backbone network in the target detection algorithm to obtain a first feature vector; performing feature sampling processing on the first feature vector by a detection head in the target detection algorithm to obtain a second feature vector; and processing the second feature vector by a detection layer in the target detection algorithm to obtain the multiple video frames.

[0012] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a fire information detection device is provided. The device comprises: an acquisition unit, used to acquire target video data of a target area; a processing unit, used to process each video frame in the target video data by a target detection algorithm to obtain multiple video frames, wherein the video frames of the multiple video frames include rectangular boxes added by the target detection algorithm, and the probability that a fire exists at a position corresponding to the rectangular box in each target video frame is greater than a preset threshold; a judgment unit, used to judge whether there is a fire in the target area based on the multiple video frames; and a determination unit, used to determine the location information of the fire based on the multiple video frames when there is a fire in the target area.

[0013] Furthermore, the judgment unit includes: a first processing subunit, used to obtain a first eigenvalue for a first video frame based on color information of an area corresponding to a rectangular box in the first video frame, wherein the first video frame is any one of the multiple video frames; a second processing subunit, used to obtain a second eigenvalue based on area information of an area corresponding to the rectangular box in the first video frame; a calculation subunit, used to calculate the similarity between the area corresponding to the rectangular box in the first video frame and the areas corresponding to the rectangular boxes in multiple second video frames, to obtain a third eigenvalue, wherein the multiple second video frames are continuous with the first video frame in a time dimension; a third processing subunit, used to obtain a fourth eigenvalue based on shape information of the area corresponding to the rectangular box in the first video frame; a judgment subunit, used to judge whether there is a fire in the target area based on the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue.

[0014] Furthermore, the first processing subunit includes: a first calculation module, used to calculate the three primary colors corresponding to each pixel point in the first area to obtain a red value, a green value and a blue value, wherein the first area is the area corresponding to the rectangular box in the first video frame; a second calculation module, used to calculate based on the red value, the green value and the blue value to obtain a first red proportion value corresponding to the first area; a first determination module, used to determine the multiple second video frames, and obtain the second red proportion value of the area corresponding to the rectangular box in each second video frame; a third calculation module, used to calculate the red proportion growth rate based on the first red proportion value and the second red proportion value to obtain the first eigenvalue.

[0015] Furthermore, the second processing sub-unit includes: a fourth calculation module, used to calculate the area of ​​the area corresponding to the rectangular box in the first video frame to obtain a first area value; a second determination module, used to determine the multiple second video frames and obtain the second area value of the area corresponding to the rectangular box in each second video frame; a fifth calculation module, used to calculate the area growth rate based on the first area value and the second area value to obtain the second eigenvalue.

[0016] Furthermore, the third processing sub-unit includes: a fifth calculation module, used to calculate the area of ​​the first region to obtain a first area value, wherein the first region is the region corresponding to the rectangular box in the first video frame; a sixth calculation module, used to calculate the perimeter of the first region to obtain a perimeter value; and a seventh calculation module, used to calculate based on the first area value and the perimeter value to obtain the fourth eigenvalue.

[0017] Furthermore, the judgment subunit includes: a judgment module, used to judge whether the first characteristic value, the second characteristic value, the third characteristic value and the fourth characteristic value meet the preset conditions; a third determination module, used to determine that if the preset conditions are met, there is a fire in the target area; and a fourth determination module, used to determine that if the preset conditions are not met, there is no fire in the target area.

[0018] Furthermore, the processing unit includes: an extraction subunit, used to perform feature extraction on each video frame through the backbone network in the target detection algorithm to obtain a first feature vector; a sampling subunit, used to perform feature sampling processing on the first feature vector through the detection head in the target detection algorithm to obtain a second feature vector; a fourth processing subunit, used to process the second feature vector through the detection layer in the target detection algorithm to obtain the multiple video frames.

[0019] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned methods for detecting fire information.

[0020] In order to achieve the above-mentioned purpose, according to another aspect of the present application, an electronic device is provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes any one of the above-mentioned methods for detecting fire information when running.

[0021] In an embodiment of the present application, target video data of a target area is acquired; each video frame in the target video data is processed by a target detection algorithm to obtain a plurality of video frames, wherein the video frames of the plurality of video frames include a rectangular box added by the target detection algorithm, and the probability that a fire exists at a position corresponding to the rectangular box in each video frame is greater than a preset threshold; based on the plurality of video frames, it is determined whether there is a fire in the target area; in the case that there is a fire in the target area, the fire location information is determined based on the plurality of video frames, thereby solving the technical problem in the related art of using temperature sensors, smoke sensors and the like for fire detection, resulting in relatively low accuracy of fire detection.

[0022] In this solution, target video data to be detected for the existence of a fire is obtained, and a corresponding rectangular frame is added to the position in each video frame where the probability of the existence of a fire is greater than a preset threshold through a target detection algorithm. Then, it is judged whether there is a fire in the target area based on multiple video frames. When it is determined that there is a fire, the fire location information is determined based on the rectangular frames in the multiple video frames. Compared with the prior art of fire detection through temperature sensors, smoke sensors, etc., it can detect whether there is a fire in the video frame more timely and accurately, thereby achieving the effect of improving the accuracy of fire detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0024] Figure 1 A hardware structure block diagram of a computer terminal for implementing a method for detecting fire information is shown;

[0025] Figure 2 The process of the fire information detection method provided in the embodiment of the present application is as follows Figure 1 ;

[0026] Figure 3 is a schematic diagram of a target detection algorithm provided according to an embodiment of the present application;

[0027] Figure 4 The process of the fire information detection method provided in the embodiment of the present application is as follows Figure 2 ;

[0028] Figure 5 is a schematic diagram of a fire information detection device provided according to an embodiment of the present application;

[0029] Figure 6 It is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0031] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0032] It should be noted that the collected information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this application are information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data are in compliance with relevant laws, regulations and standards, necessary confidentiality measures are taken, and public order and good customs are not violated, and corresponding operation entrances are provided for users to choose to authorize or refuse. For example, an interface is set up between this system and relevant users or institutions to provide users with corresponding operation entrances for users to choose to agree or refuse the results of automated decision-making; if the user chooses to refuse, the expert decision-making process will be entered.

[0033] Example 1

[0034] According to an embodiment of the present application, a method embodiment for detecting fire information is also provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0035] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 FIG. 1 shows a hardware structure block diagram of a computer terminal (or mobile device) for implementing a method for detecting fire information. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more (102a, 102b, ..., 102n are used to illustrate) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0036] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10 (or mobile device). As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0037] The memory 104 can be used to store software programs and modules of application software, such as program instructions / data storage devices corresponding to the fire information detection method in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, to implement the above-mentioned fire information detection method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0038] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0039] The display may be a touch screen type liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0040] Under the above operating environment, this application provides Figure 2 The fire information detection method shown. Figure 2 The process of the fire information detection method according to the first embodiment of the present application is as follows Figure 1 .

[0041] Step S201, obtaining target video data of a target area.

[0042] Optionally, target video data of the target area may be acquired in real time through acquisition devices such as cameras. It should be noted that the target area may be a branch of a financial institution or a large computer room, or an outdoor environment.

[0043] Step S202, processing each video frame in the target video data by a target detection algorithm to obtain multiple video frames, wherein the video frames of the multiple video frames include rectangular boxes added by the target detection algorithm, and the probability of the existence of a fire at a position corresponding to the rectangular box in each target video frame is greater than a preset threshold.

[0044] Optionally, after obtaining the above-mentioned target video data, each video frame in the target video data is processed by a target detection algorithm. It should be noted that the target detection algorithm can be a YOLO algorithm. The YOLO algorithm can process more than 45 frames of monitoring images per second. When a fire occurs, it can quickly identify and use a box to frame the fire range. Therefore, each video frame in the target video data is detected by the target detection algorithm, and the area where the probability of fire is greater than a preset threshold is processed by rectangular box selection, so as to obtain the above-mentioned multiple video frames.

[0045] Step S203: determining whether there is a fire in the target area based on the multiple video frames.

[0046] Optionally, after multiple video frames with rectangular frames, it is further determined whether there is a fire in the target area based on these video frames, for example, based on pixel information of the area corresponding to the rectangular frame in the multiple video frames, it is further determined whether there is a fire in the target area.

[0047] Step S204: when there is a fire in the target area, determine the location information of the fire based on the multiple video frames.

[0048] Optionally, when it is determined that there is a fire in the target area, the location information of the fire is determined based on the area of ​​the rectangular frame where the fire exists in the multiple video frames. In an optional embodiment, the area where the fire exists in the multiple video frames can be visualized to prompt the user to process.

[0049] To summarize, the target video data to be detected for the existence of a fire is obtained, and a corresponding rectangular box is added to the position in each video frame where the probability of the existence of a fire is greater than a preset threshold through a target detection algorithm, and then a judgment is made based on multiple video frames whether there is a fire in the target area. When it is determined that there is a fire, the fire location information is determined based on the rectangular boxes in the multiple video frames. Compared with the fire detection through temperature sensors, smoke sensors and other detectors in the prior art, it can detect whether there is a fire in the video frame more timely and accurately, thereby achieving the effect of improving the accuracy of fire detection.

[0050] Optionally, in the fire information detection method provided in the embodiment of the present application, judging whether there is a fire in the target area based on multiple video frames includes: for a first video frame, obtaining a first eigenvalue based on color information of an area corresponding to a rectangular box in the first video frame, wherein the first video frame is any one of the multiple video frames; obtaining a second eigenvalue based on area information of an area corresponding to the rectangular box in the first video frame; calculating the similarity between the area corresponding to the rectangular box in the first video frame and the area corresponding to the rectangular box in multiple second video frames to obtain a third eigenvalue, wherein the multiple second video frames are continuous with the first video frame in the time dimension; obtaining a fourth eigenvalue based on shape information of the area corresponding to the rectangular box in the first video frame; judging whether there is a fire in the target area based on the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue.

[0051] In an optional embodiment, judging whether there is a fire in a target area through multiple video frames includes the following steps: for any one of the multiple video frames, determining the area corresponding to the rectangular box in the first video frame, and obtaining a first eigenvalue based on the color information of the area. Flame color is an important feature of the flame. Through the analysis of a large number of flame images, it is found that the color of the flame image has a certain regularity: in the RGB color space, R channel>G channel>B channel; the color of the flame tends to move from white to red from the core to the outer flame; the color of the flame is easily affected by light, and it appears as different colors in different environments. Therefore, the first eigenvalue can be obtained based on the color information of the area.

[0052] Then, the second eigenvalue is calculated based on the area information of the area corresponding to the rectangular frame in the first video frame. Since the shape of the flame is irregular when observed from a single frame image, but has a certain similarity when observed from a sequence of images, especially continuous frame images with a short interval, the shape of the flame can be calculated to obtain a third eigenvalue by calculating the similarity between the area corresponding to the rectangular frame in the first video frame and the area corresponding to the rectangular frame in the plurality of second video frames. It should be noted that the plurality of second video frames are continuous with the first video frame in the time dimension, for example, 5 continuous frames are used as the second video frame.

[0053] In an optional embodiment, the similarity calculation formula of adjacent frames is as follows:

[0054]

[0055] Where i=1, 2...N, i is the image frame number, (x, y) is the position of each pixel in the image, A is the suspected area, k=1, 2...M, k is the label of the suspected area of ​​the image.

[0056] The edge of the flame of an early fire is constantly changing and irregular due to the influence of external factors and fire plumes, while the edge contours of other incandescent lamps, candles, and neon lights on the road that have similar color characteristics to flames are more regular. Therefore, the fourth eigenvalue can be obtained based on the shape information of the area corresponding to the rectangular frame in the first video frame.

[0057] Finally, it is determined whether there is a fire in the target area based on the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue.

[0058] The corresponding eigenvalues ​​are obtained through color, area, similarity and shape information, and whether there is a fire in the target area is judged based on the eigenvalues, which can effectively improve the accuracy of fire detection.

[0059] Optionally, in the fire information detection method provided in the embodiment of the present application, obtaining the first eigenvalue based on the color information of the area corresponding to the rectangular box in the first video frame includes: calculating the three primary colors corresponding to each pixel point in the first area to obtain a red value, a green value and a blue value, wherein the first area is the area corresponding to the rectangular box in the first video frame; calculating based on the red value, the green value and the blue value to obtain a first red proportion value corresponding to the first area; determining multiple second video frames, and obtaining a second red proportion value of the area corresponding to the rectangular box in each second video frame; and calculating the red proportion growth rate based on the first red proportion value and the second red proportion value to obtain the first eigenvalue.

[0060] In an optional embodiment, the following steps may be used to calculate the first eigenvalue: determine the area corresponding to the rectangular box in the first video frame, i.e., the first area mentioned above, the color of the flame image in the RGB color space, R channel>G channel>B channel; the color of the flame tends to move from white to red from the core to the outer flame, therefore, the red value, green value and blue value, i.e., RGB value, may be calculated based on the three primary colors corresponding to each pixel point in the first area, and then the first red proportion value corresponding to the first area may be obtained based on the red value, green value and blue value.

[0061] For example, the first red specific gravity value is calculated using the following formula:

[0062]

[0063] Among them, R(p(x, y)), G(p(x, y)), and B(p(x, y)) are the R, G, and B values ​​of the pixel p(x, y) in area A, respectively.

[0064] After obtaining the above-mentioned first red proportion value, in order to avoid the problem of error in the red proportion value of a single frame, the second red proportion value of the area corresponding to the rectangular box in multiple second video frames is calculated, and then the red proportion growth rate between the first red proportion value and the second red proportion value is calculated, and finally the red proportion growth rate is determined as the above-mentioned first characteristic value.

[0065] In an optional embodiment, the red weight growth rate between two adjacent frames is: In order to improve the accuracy of calculation, the average growth rate R of the red proportion between 5 consecutive frames is taken as the first eigenvalue mentioned above.

[0066] The flame color distribution law is described by the red proportion growth rate to improve the accuracy of subsequent fire detection.

[0067] Optionally, in the fire information detection method provided in the embodiment of the present application, obtaining the second eigenvalue based on the area information of the area corresponding to the rectangular box in the first video frame includes: calculating the area of ​​the area corresponding to the rectangular box in the first video frame to obtain a first area value; determining multiple second video frames, and obtaining the second area value of the area corresponding to the rectangular box in each second video frame; and calculating the area growth rate based on the first area value and the second area value to obtain a second eigenvalue.

[0068] In an optional embodiment, the second eigenvalue is calculated by the following steps: the area of ​​the region corresponding to the rectangular box in the first video frame is calculated to obtain a first area value, and the second area value of the region corresponding to the rectangular box in multiple consecutive frames (i.e., the second video frame mentioned above) is obtained, and the area growth rate is calculated based on the first area value and the second area value to obtain the second eigenvalue.

[0069] In an optional embodiment, let t i The area of ​​the kth region at time is A i，k , t i+j The area of ​​the kth region at time is A i+j，k In video images, time is usually divided by the number of frames, so it can be considered that S i It is the target area A of the i-th frame i，k The area, S i+j，k It is the k-th region A of the (i+j)th frame i+j，k The area growth rate is calculated as follows:

[0070]

[0071] In practical applications, the area growth rate of five consecutive frames can be calculated, which can eliminate the interference of some fixed light sources and fast-moving objects and improve the recognition rate of subsequent fire detection.

[0072] Optionally, in the fire information detection method provided in the embodiment of the present application, obtaining the fourth eigenvalue based on the shape information of the area corresponding to the rectangular box in the first video frame includes: calculating the area of ​​the first area to obtain a first area value, wherein the first area is the area corresponding to the rectangular box in the first video frame; calculating the perimeter of the first area to obtain a perimeter value; and calculating based on the first area value and the perimeter value to obtain the fourth eigenvalue.

[0073] In an optional embodiment, the fourth characteristic information is calculated by the following steps: a first area value is obtained by calculating the area of ​​the area corresponding to the rectangular frame in the first video frame, a perimeter of the area is calculated to obtain a perimeter value, and a fourth characteristic value is calculated based on the first area value and the perimeter value.

[0074] In an optional embodiment, calculating the fourth eigenvalue according to the first area value and the perimeter value includes calculating the circularity according to the first area value and the perimeter value to obtain the fourth eigenvalue.

[0075] In an optional embodiment, the calculation formula of the fourth eigenvalue is as follows:

[0076]

[0077] Among them, A i，k represents the area of ​​the kth region in the current i-th frame, C i，k is the perimeter of the contour of the kth region in the i-th frame.

[0078] The circularity can accurately describe the edge contour shape characteristics of the flame and improve the accuracy of fire judgment.

[0079] Optionally, in the fire information detection method provided in the embodiment of the present application, judging whether there is a fire in the target area based on the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue includes: judging whether the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue meet a preset condition; if the preset condition is met, there is a fire in the target area; if the preset condition is not met, there is no fire in the target area.

[0080] In an optional embodiment, judging whether there is a fire in the target area includes the following steps: corresponding thresholds can be set for the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue, respectively, and then the magnitude relationship between the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue and the corresponding thresholds is judged. If the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue are all greater than or equal to the corresponding thresholds, then the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue are judged to meet the preset conditions, that is, there is a fire in the first area corresponding to the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue. If the preset conditions are not met, it means that there is no fire in the first area corresponding to the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue. It should be noted that for each rectangular box in each video frame, the corresponding eigenvalue needs to be calculated in the above manner, and the eigenvalue is judged to determine whether there is a fire in the area based on the eigenvalue.

[0081] The above-mentioned first eigenvalue, second eigenvalue, third eigenvalue and fourth eigenvalue help to perform refined analysis and filtering on the detection results of the target detection algorithm, further improve the accuracy of fire source detection and reduce the false alarm rate.

[0082] Optionally, in the fire information detection method provided in the embodiment of the present application, each video frame in the target video data is processed by a target detection algorithm to obtain multiple video frames, including: performing feature extraction on each video frame by a backbone network in the target detection algorithm to obtain a first feature vector; performing feature sampling processing on the first feature vector by a detection head in the target detection algorithm to obtain a second feature vector; and processing the second feature vector by a detection layer in the target detection algorithm to obtain multiple video frames.

[0083] In an optional embodiment, the schematic diagram of the target detection algorithm is as follows: Figure 3As shown in the figure, the target detection algorithm consists of four parts: Input, Backbone, Head, and Detect. The Input is a 640*640*3 data input. Backbone is the backbone network composed of CBS, ELAN, and MP-1. Head consists of CBS, SPPCSPC, E-ELAN, MP-2, and RepConv. CBS consists of three parts, which are two-dimensional convolution, batch normalization, and SiLU activation function. The ELAN module consists of the longest and shortest gradient paths. More blocks are stacked through the shortest path to learn more features. The overall role of MP: to achieve downsampling to reduce feature loss. The MP-1 module consists of 2 branches, and MP-2 consists of 3 branches. The first branch first uses Maxpool (maximum pooling) to achieve downsampling, and then uses a 1*1 convolution to achieve the change in the number of channels. The other branch of MP-1 is a 3*3 convolution kernel with a stride of 2, which achieves downsampling. The function of SPP is to achieve the fusion of information of different feature scales. It uses four different scales of maximum pooling for processing. The pooling kernel sizes of the maximum pooling are 13x13, 9x9, 5x5, and 1x1 (1x1 means no processing). The CSP module consists of two parts, one part is processed by SPP structure, and the other part is processed by 1*1 convolution for channel number processing, and finally the two parts are concat. SPP CSP achieves the fusion of information of different feature scales, reduces the amount of calculation, and improves the speed. The last part of the head is RepConv. The training stage consists of multiple branches, 3x3, 1x1, and identity (mapping). The inference stage becomes only one 3x3 convolution, which reduces the number of parameters and speeds up the inference speed. RepConv learns more features during the training process, and speeds up the inference process. Detect generates soft labels through the optimizer based on the results of the Lead head and GT (true value of the label). Soft labels are used when training models with the Aux head and the Lead head. The reason for this is that the Lead head has a relatively strong learning ability, so the soft labels generated by it should be more representative of the distribution and correlation between the source data and the target data. In addition, residual learning is added to allow the shallower Aux head to directly learn the information learned by the Lead head, so that the Lead head can better focus on learning the residual information that has not been learned.

[0084] Therefore, each video frame in the target video data is processed by the target detection algorithm to obtain multiple video frames, including: performing feature extraction on each video frame through Backbone (i.e. the above-mentioned backbone network) to obtain a first feature vector, performing feature sampling processing on the first feature vector through Head (i.e. the above-mentioned detection head) to obtain a second feature vector, and finally processing the second feature vector through Detect (i.e. the above-mentioned detection layer) to obtain multiple video frames.

[0085] The target detection algorithm can detect multiple targets in one image at the same time, including small fire sources, thereby improving the comprehensiveness of detection.

[0086] In an optional embodiment, the following may be used: Figure 4 The flowchart shown realizes the detection of fire: obtaining the video data to be detected, detecting the video data through YOLO7, and using a box to frame the fire range, and finally, determining whether there is a fire source based on the video frame with the box selection.

[0087] The fire information detection method provided in the embodiment of the present application obtains target video data of a target area; processes each video frame in the target video data through a target detection algorithm to obtain multiple video frames, wherein the video frames of the multiple video frames include rectangular boxes added by the target detection algorithm, and the probability that a fire exists at a position corresponding to the rectangular box in each video frame is greater than a preset threshold; determines whether there is a fire in the target area based on the multiple video frames; and when there is a fire in the target area, determines the fire location information based on the multiple video frames, thereby solving the technical problem in the related art that fire detection is performed through temperature sensors, smoke sensors, and the like, resulting in relatively low accuracy of fire detection.

[0088] In this solution, target video data to be detected for the existence of a fire is obtained, and a corresponding rectangular frame is added to the position in each video frame where the probability of the existence of a fire is greater than a preset threshold through a target detection algorithm. Then, it is judged whether there is a fire in the target area based on multiple video frames. When it is determined that there is a fire, the fire location information is determined based on the rectangular frames in the multiple video frames. Compared with the prior art of fire detection through temperature sensors, smoke sensors, etc., it can detect whether there is a fire in the video frame more timely and accurately, thereby achieving the effect of improving the accuracy of fire detection.

[0089] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0090] Example 2

[0091] The embodiment of the present application also provides a fire information detection device. It should be noted that the fire information detection device of the embodiment of the present application can be used to execute the fire information detection method provided in the embodiment of the present application. The fire information detection device provided in the embodiment of the present application is introduced below.

[0092] According to an embodiment of the present application, a device for implementing the above-mentioned fire information detection method is also provided, such as Figure 5 As shown, the device includes: an acquisition unit 501, a processing unit 502, a judgment unit 503 and a determination unit 504.

[0093] An acquisition unit 501 is used to acquire target video data of a target area;

[0094] The processing unit 502 is used to process each video frame in the target video data by using the target detection algorithm to obtain multiple video frames, wherein the video frames of the multiple video frames include a rectangular frame added by the target detection algorithm, and the probability that a fire exists at a position corresponding to the rectangular frame in each target video frame is greater than a preset threshold;

[0095] A determination unit 503, configured to determine whether there is a fire in the target area based on a plurality of video frames;

[0096] The determination unit 504 is configured to determine the location information of the fire based on a plurality of video frames when there is a fire in the target area.

[0097] The fire information detection device provided in the embodiment of the present application obtains the target video data of the target area through the acquisition unit 501; the processing unit 502 processes each video frame in the target video data through the target detection algorithm to obtain multiple video frames, wherein the video frames of the multiple video frames include rectangular boxes added by the target detection algorithm, and the probability that a fire exists at the position corresponding to the rectangular box in each target video frame is greater than a preset threshold; the judgment unit 503 judges whether there is a fire in the target area based on the multiple video frames; the determination unit 504 determines the location information of the fire based on the multiple video frames when there is a fire in the target area, thereby solving the technical problem in the related art that the accuracy of fire detection is relatively low due to the use of temperature-sensing, smoke-sensing and other detectors for fire detection.

[0098] In this solution, target video data to be detected for the existence of a fire is obtained, and a corresponding rectangular frame is added to the position in each video frame where the probability of the existence of a fire is greater than a preset threshold through a target detection algorithm. Then, it is judged whether there is a fire in the target area based on multiple video frames. When it is determined that there is a fire, the fire location information is determined based on the rectangular frames in the multiple video frames. Compared with the prior art of fire detection through temperature sensors, smoke sensors, etc., it can detect whether there is a fire in the video frame more timely and accurately, thereby achieving the effect of improving the accuracy of fire detection.

[0099] Optionally, in the fire information detection device provided in the embodiment of the present application, the judgment unit includes: a first processing subunit, used to obtain a first eigenvalue for a first video frame based on color information of an area corresponding to a rectangular box in the first video frame, wherein the first video frame is any one of multiple video frames; a second processing subunit, used to obtain a second eigenvalue based on area information of an area corresponding to the rectangular box in the first video frame; a calculation subunit, used to calculate the similarity between the area corresponding to the rectangular box in the first video frame and the areas corresponding to the rectangular boxes in multiple second video frames, to obtain a third eigenvalue, wherein the multiple second video frames are continuous with the first video frame in the time dimension; a third processing subunit, used to obtain a fourth eigenvalue based on shape information of the area corresponding to the rectangular box in the first video frame; a judgment subunit, used to judge whether there is a fire in the target area based on the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue.

[0100] Optionally, in the fire information detection device provided in the embodiment of the present application, the first processing subunit includes: a first calculation module, used to calculate the three primary colors corresponding to each pixel point in the first area to obtain a red value, a green value and a blue value, wherein the first area is the area corresponding to the rectangular box in the first video frame; a second calculation module, used to calculate based on the red value, the green value and the blue value to obtain a first red proportion value corresponding to the first area; a first determination module, used to determine multiple second video frames, and obtain the second red proportion value of the area corresponding to the rectangular box in each second video frame; a third calculation module, used to calculate the red proportion growth rate based on the first red proportion value and the second red proportion value to obtain a first eigenvalue.

[0101] Optionally, in the fire information detection device provided in the embodiment of the present application, the second processing subunit includes: a fourth calculation module, used to calculate the area of ​​the area corresponding to the rectangular box in the first video frame to obtain a first area value; a second determination module, used to determine multiple second video frames and obtain the second area value of the area corresponding to the rectangular box in each second video frame; a fifth calculation module, used to calculate the area growth rate based on the first area value and the second area value to obtain a second eigenvalue.

[0102] Optionally, in the fire information detection device provided in the embodiment of the present application, the third processing subunit includes: a fifth calculation module, used to calculate the area of ​​the first area to obtain a first area value, wherein the first area is the area corresponding to the rectangular frame in the first video frame; a sixth calculation module, used to calculate the perimeter of the first area to obtain a perimeter value; and a seventh calculation module, used to perform calculation based on the first area value and the perimeter value to obtain a fourth eigenvalue.

[0103] Optionally, in the fire information detection device provided in the embodiment of the present application, the judgment subunit includes: a judgment module, used to judge whether the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue meet the preset conditions; a third determination module, used to determine that if the preset conditions are met, there is a fire in the target area; and a fourth determination module, used to determine that if the preset conditions are not met, there is no fire in the target area.

[0104] Optionally, in the fire information detection device provided in the embodiment of the present application, the processing unit includes: an extraction subunit, used to extract features from each video frame through a backbone network in a target detection algorithm to obtain a first feature vector; a sampling subunit, used to perform feature sampling processing on the first feature vector through a detection head in the target detection algorithm to obtain a second feature vector; and a fourth processing subunit, used to process the second feature vector through a detection layer in the target detection algorithm to obtain multiple video frames.

[0105] It should be noted that the acquisition unit 501, processing unit 502, judgment unit 503 and determination unit 504 described above correspond to steps S201 to S204 in Embodiment 1, and the four units and corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in Embodiment 1 described above. It should be noted that the modules or units described above may be hardware components or software components stored in a memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n), and the modules described above may also be run in the computer terminal 10 provided in Embodiment 1 as part of the device.

[0106] Example 3

[0107] An embodiment of the present application may provide an electronic device, Figure 6 is a structural block diagram of an electronic device according to an embodiment of the present application. Figure 6 As shown, the electronic device may include: one or more ( Figure 6 (only one is shown) processor 602, memory 604, storage controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0108] Among them, the memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, realizing the above-mentioned method. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0109] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: obtain target video data of the target area; process each video frame in the target video data through the target detection algorithm to obtain multiple video frames, wherein the video frames of the multiple video frames include rectangular boxes added by the target detection algorithm, and the probability that a fire exists at a position corresponding to the rectangular box in each target video frame is greater than a preset threshold; based on the multiple video frames, determine whether there is a fire in the target area; if there is a fire in the target area, determine the location information of the fire based on the multiple video frames.

[0110] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: judging whether there is a fire in the target area based on multiple video frames includes: for the first video frame, obtaining a first eigenvalue based on the color information of the area corresponding to the rectangular box in the first video frame, wherein the first video frame is any one of the multiple video frames; obtaining a second eigenvalue based on the area information of the area corresponding to the rectangular box in the first video frame; calculating the similarity between the area corresponding to the rectangular box in the first video frame and the area corresponding to the rectangular box in multiple second video frames to obtain a third eigenvalue, wherein the multiple second video frames are continuous with the first video frame in the time dimension; obtaining a fourth eigenvalue based on the shape information of the area corresponding to the rectangular box in the first video frame; judging whether there is a fire in the target area based on the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue.

[0111] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: based on the color information of the area corresponding to the rectangular box in the first video frame, obtain the first eigenvalue including: calculating the three primary colors corresponding to each pixel point in the first area to obtain the red value, green value and blue value, wherein the first area is the area corresponding to the rectangular box in the first video frame; calculating based on the red value, green value and blue value to obtain the first red proportion value corresponding to the first area; determining multiple second video frames, and obtaining the second red proportion value of the area corresponding to the rectangular box in each second video frame; calculating the red proportion growth rate based on the first red proportion value and the second red proportion value to obtain the first eigenvalue.

[0112] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: based on the area information of the area corresponding to the rectangular box in the first video frame, obtain the second eigenvalue including: calculating the area of ​​the area corresponding to the rectangular box in the first video frame to obtain the first area value; determining multiple second video frames, and obtaining the second area value of the area corresponding to the rectangular box in each second video frame; calculating the area growth rate based on the first area value and the second area value to obtain the second eigenvalue.

[0113] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: obtaining the fourth eigenvalue based on the shape information of the area corresponding to the rectangular box in the first video frame includes: calculating the area of ​​the first area to obtain a first area value, wherein the first area is the area corresponding to the rectangular box in the first video frame; calculating the perimeter of the first area to obtain a perimeter value; and calculating based on the first area value and the perimeter value to obtain the fourth eigenvalue.

[0114] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: judging whether there is a fire in the target area based on the first characteristic value, the second characteristic value, the third characteristic value and the fourth characteristic value, including: judging whether the first characteristic value, the second characteristic value, the third characteristic value and the fourth characteristic value meet the preset conditions; if the preset conditions are met, there is a fire in the target area; if the preset conditions are not met, there is no fire in the target area.

[0115] The processor can call the information and application programs stored in the memory through the transmission device to perform the following steps: processing each video frame in the target video data through the target detection algorithm to obtain multiple video frames, including: extracting features from each video frame through the backbone network in the target detection algorithm to obtain a first feature vector; performing feature sampling processing on the first feature vector through the detection head in the target detection algorithm to obtain a second feature vector; processing the second feature vector through the detection layer in the target detection algorithm to obtain multiple video frames.

[0116] It can be understood by those skilled in the art that Figure 6 The structure shown is for illustration only, and the electronic device may also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, a mobile Internet device (Mobile Internet Devices, MID), a PAD, or other terminal devices. Figure 6 The structure of the electronic device is not limited. Figure 6 More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Figure 6 Different configurations are shown.

[0117] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0118] Example 4

[0119] The embodiment of the present application further provides a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the fire information detection method provided in the first embodiment.

[0120] Optionally, in this embodiment, the above storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0121] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing the program steps of the method for detecting fire information.

[0122] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0123] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0124] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0125] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0126] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0127] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0128] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for detecting fire information, characterized in that: include: Acquire target video data of the target area; Processing each video frame in the target video data by a target detection algorithm to obtain a plurality of video frames, wherein the video frames of the plurality of video frames include a rectangular frame added by the target detection algorithm, and a probability that a fire exists at a position corresponding to the rectangular frame in each target video frame is greater than a preset threshold; Determining whether there is a fire in the target area based on the multiple video frames; In the case that there is a fire in the target area, location information of the fire is determined based on the multiple video frames.

2. The method according to claim 1, characterized in that Judging whether there is a fire in the target area according to the multiple video frames includes: For a first video frame, obtaining a first feature value according to color information of an area corresponding to a rectangular frame in the first video frame, wherein the first video frame is any one of the multiple video frames; Obtaining a second eigenvalue according to area information of a region corresponding to the rectangular frame in the first video frame; Calculating similarities between an area corresponding to a rectangular frame in the first video frame and areas corresponding to rectangular frames in a plurality of second video frames to obtain a third eigenvalue, wherein the plurality of second video frames are continuous with the first video frame in a time dimension; Obtaining a fourth eigenvalue according to shape information of an area corresponding to the rectangular frame in the first video frame; Whether there is a fire in the target area is determined based on the first eigenvalue, the second eigenvalue, the third eigenvalue and the fourth eigenvalue.

3. The method according to claim 2, characterized in that Obtaining the first characteristic value according to the color information of the area corresponding to the rectangular frame in the first video frame includes: Calculate the three primary colors corresponding to each pixel point in the first area to obtain a red value, a green value, and a blue value, wherein the first area is an area corresponding to the rectangular frame in the first video frame; Calculating according to the red value, the green value and the blue value to obtain a first red specific gravity value corresponding to the first area; Determine the plurality of second video frames, and obtain a second red proportion value of an area corresponding to a rectangular frame in each second video frame; The red specific gravity growth rate is calculated according to the first red specific gravity value and the second red specific gravity value to obtain the first characteristic value.

4. The method according to claim 2, characterized in that: The second eigenvalue is obtained according to the area information of the region corresponding to the rectangular frame in the first video frame, including: Calculating the area of ​​a region corresponding to the rectangular frame in the first video frame to obtain a first area value; Determine the plurality of second video frames, and obtain a second area value of a region corresponding to a rectangular frame in each second video frame; The area growth rate is calculated according to the first area value and the second area value to obtain the second characteristic value.

5. The method according to claim 2, characterized in that: The fourth eigenvalue obtained according to the shape information of the area corresponding to the rectangular frame in the first video frame includes: Calculating an area of ​​a first region to obtain a first area value, wherein the first region is a region corresponding to a rectangular frame in the first video frame; Calculating the perimeter of the first area to obtain a perimeter value; The fourth eigenvalue is obtained by performing calculation according to the first area value and the perimeter value.

6. The method according to claim 2, characterized in that Judging whether there is a fire in the target area according to the first characteristic value, the second characteristic value, the third characteristic value, and the fourth characteristic value includes: Determine whether the first characteristic value, the second characteristic value, the third characteristic value and the fourth characteristic value meet a preset condition; If the preset condition is met, there is a fire in the target area; If the preset condition is not met, there is no fire in the target area.

7. The method according to claim 1, characterized in that Each video frame in the target video data is processed by a target detection algorithm to obtain a plurality of video frames including: Extract features from each video frame using the backbone network in the target detection algorithm to obtain a first feature vector; Performing feature sampling processing on the first feature vector by a detection head in the target detection algorithm to obtain a second feature vector; The second feature vector is processed by the detection layer in the target detection algorithm to obtain the multiple video frames.

8. A fire information detection device, characterized in that: include: An acquisition unit, used for acquiring target video data of a target area; a processing unit, configured to process each video frame in the target video data by using a target detection algorithm to obtain a plurality of video frames, wherein the video frames of the plurality of video frames include a rectangular frame added by the target detection algorithm, and a probability that a fire exists at a position corresponding to the rectangular frame in each target video frame is greater than a preset threshold; A judging unit, configured to judge whether there is a fire in the target area according to the plurality of video frames; The determination unit is used to determine the location information of the fire according to the multiple video frames when there is a fire in the target area.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is executed, the device where the computer-readable storage medium is located is controlled to execute the fire information detection method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: A memory storing an executable program; A processor is used to run the program, wherein the program executes the fire information detection method according to any one of claims 1 to 7 when running.

Images