A fire fault identification method and device based on multiple factors

Abstract

The embodiment of the present invention provides a fire fault diagnosis method and device based on multiple factors, which relates to the technical field of fire fault identification technology. The method includes: photographing the monitored area through a pre-set camera to obtain a number of images; extracting and comparing the features of the images, performing a first comparison between the feature values ​​of each image obtained and a plurality of pre-determined fault image feature thresholds, and then performing a second comparison between the feature values ​​of the currently obtained image and the feature values ​​of the image at the next moment in the same cycle; performing a first fault identification based on the results of the two comparisons; in addition, real-time environmental information is collected through multiple sensing nodes, and a second fault identification is performed using the real-time environmental information and preset safety environmental information; based on the first fault identification result and the second fault identification, determining whether the fire equipment has a fault. Through the present invention, the problem of low accuracy in fire fault identification is solved, thereby achieving the effect of improving the accuracy of fire fault identification.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of fire fault identification, and in particular, to a fire fault identification method and device based on multiple factors. Background Art

[0002] Fire protection system equipment may malfunction during daily operation. The main causes of equipment failure include failure of the equipment's own components, functional failure caused by dust accumulation, disassembly during equipment repair or maintenance, and disconnection of communication lines during equipment repair or maintenance.

[0003] At present, the fire protection system is generally monitored by monitoring the temperature, water pressure and other data during operation, and the monitored data is compared with the preset threshold. When the value of the monitored data exceeds the preset threshold, it is determined that the battery has failed. However, in real scenarios, there are influencing factors such as noise during the operation of the fire protection system, which can easily cause misjudgment and lead to a low recognition rate of fire protection failures.

[0004] There is currently a lack of better solutions to the above problems. Summary of the invention

[0005] The embodiment of the present invention provides a method and device for identifying faults of a fire protection system based on multiple factors, so as to at least solve the problem of low accuracy of fire protection fault diagnosis in the related art.

[0006] According to one embodiment of the present invention, a fault identification method for a fire protection system based on multiple factors is provided, comprising: acquiring images of an area to be monitored, shooting with a preset camera according to a preset period and a preset camera rotation route to acquire a plurality of images; performing feature extraction on the plurality of images, extracting feature values ​​of each of the plurality of images based on a preset algorithm; performing a first comparison between the feature values ​​of each of the acquired images and a plurality of predetermined fault image feature thresholds, and performing a second comparison between the feature values ​​of each of the acquired images and feature values ​​of an image at the next moment in the same period; performing a first fault identification based on the first comparison result and the second comparison result; collecting environmental information of the area to be monitored, collecting real-time environmental information through a plurality of sensing nodes arranged in the area to be monitored, and performing a third comparison between the real-time environmental information and preset safety environmental information, and performing a second fault identification based on the third comparison result; determining whether a fire protection equipment has a fault based on the first fault identification result and the second fault identification result.

[0007] Preferably, the performing feature extraction on the plurality of images, extracting a feature value of each of the plurality of images based on a preset algorithm, comprises:

[0008] After obtaining a number of images, the method also includes preprocessing the images, denoising the images using a Gaussian denoising method, and enhancing the image quality using an averaging method. Then, each of the images is processed based on a deep convolutional network to obtain local features of each of the images; the local features are pooled to obtain global features of the images; the local features and the global features are fused to obtain feature values ​​of each of the images.

[0009] Preferably, the first comparing the acquired feature value of each image with a plurality of predetermined fault image feature thresholds comprises:

[0010] If the first comparison result is that the feature value of each of the images is greater than or equal to any one of the fault image feature thresholds, then determining that the first comparison result is that a fire fault exists in the area to be monitored;

[0011] If the first comparison result is that the feature value of each of the images is smaller than the plurality of fault image feature thresholds, it is determined that the first comparison result is that there is no fire fault in the area to be monitored.

[0012] Preferably, the second comparison of the acquired characteristic value of each image with the characteristic value of an image at the next moment in the same period includes:

[0013] Processing the current image and the image at the next moment based on a deep convolutional network to obtain local features of the current image and local features of the image at the next moment; performing pooling processing on the local features of the current image and the local features of the image at the next moment respectively to obtain global features of the current image and global features of the image at the next moment; fusing the local features of the current image and the global features of the current image to obtain feature values ​​of the current image; fusing the local features of the image at the next moment and the global features of the image at the next moment to obtain feature values ​​of the image at the next moment;

[0014] A second comparison is performed on the feature value of the current image and the feature value of the image at the next moment to obtain a second comparison result.

[0015] Preferably, performing a second comparison on the feature value of the current image and the feature value of the image at the next moment to obtain a second comparison result includes:

[0016] Based on a deep convolutional network, convolution processing is performed on the image at the current moment and the image at the next moment of the same cycle to obtain local features of the image at the current moment and local features of the image at the next moment; pooling processing is performed on the local features of the image at the current moment and the local features of the image at the next moment of the same cycle respectively to obtain global features of the image at the current moment and global features of the image at the next moment of the same cycle; the local features of the image at the current moment and the global features of the image at the current moment are fused to obtain feature values ​​of the image at the current moment; the local features of the image at the next moment of the same cycle and the global features of the image at the next moment of the same cycle are fused to obtain feature values ​​of the image at the next moment of the same cycle;

[0017] A second comparison is performed on the feature value of the image at the current moment and the feature value of the image at the next moment in the same period to obtain a second comparison result.

[0018] Preferably, performing a first fire fault identification based on the first comparison result and the second comparison result includes:

[0019] If the first comparison result and the second comparison result both indicate that a fire fault exists in the area to be monitored, then the first fire fault identification result indicates that a fire fault exists in the area to be monitored;

[0020] If at least one of the first comparison result and the second comparison result is that the fire fault does not exist in the monitored area, then the first fire fault identification result is pending.

[0021] Preferably, real-time environment information is collected by arranging a plurality of sensing nodes in the area to be monitored, and a third comparison is performed between the real-time environment information and preset safety environment information, and a second fault identification is performed based on the third comparison result, including:

[0022] Collecting real-time environmental information by arranging a plurality of sensing nodes in the area to be monitored, wherein the plurality of sensing nodes include: a temperature sensor, a carbon dioxide sensor and a smoke sensor;

[0023] When at least one item of information collected by the temperature sensor, the carbon dioxide sensor and the two smoke sensors exceeds the preset safety environment information, the second fault identification result is determined to be that there is a fire fault in the monitored area; otherwise, it is considered that the second fault identification result is that there is no fire fault in the monitored area.

[0024] Preferably, the determining whether a fire-fighting equipment fails based on the first fault identification result and the second fault identification includes: if the first fault identification result and the second fault identification both indicate that a fire-fighting fault exists in the area to be monitored, then determining that a fire-fighting fault exists in the area to be monitored; if at least one of the first fault identification result and the second fault identification indicates that there is no fire-fighting fault or is to be determined, then determining that there is no fire-fighting fault in the area to be monitored.

[0025] According to another embodiment of the present invention, a fire fault identification device based on multiple factors is provided, including: an image acquisition module, which acquires a plurality of images by shooting with a preset camera according to a preset period and a preset camera rotation route; a feature extraction module, which extracts a feature value of each of the plurality of images based on a preset algorithm; a comparison module, which performs a first comparison between the feature value of each of the acquired images and a plurality of predetermined fault image feature thresholds, and performs a second comparison between the feature value of each of the acquired images and a feature value of an image at the next moment in the same period; a first fault identification module, which performs a first fault identification based on the first comparison result and the second comparison result; a second fault identification module, which acquires real-time environmental information through a plurality of sensing nodes arranged in the area to be detected, and performs a third comparison between the real-time environmental information and a preset safety environment, and performs a second fault identification based on the third comparison result; a fault determination module, which determines whether a fire-fighting equipment has a fault based on the first fault identification result and the second fault identification result.

[0026] According to yet another embodiment of the present invention, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the steps of any one of the above method embodiments when run.

[0027] According to yet another embodiment of the present invention, there is provided an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0028] Through the present invention, a camera to be monitored is used to shoot the area to be monitored to obtain a number of images; feature extraction and comparison are performed on the images, and the feature value of each image obtained is compared with a plurality of predetermined fault image feature thresholds for a first time, and then the feature value of the currently obtained image is compared with the feature information of the image at the next moment in the same cycle for a second time; a first fault identification is performed based on the results of the two comparisons; in addition, real-time environmental information is collected through a plurality of sensing nodes, and a second fault identification is performed using the real-time environmental information and preset safety environmental information; based on the first fault identification result and the second fault identification, it is determined whether the fire-fighting equipment has a fault. The present invention uses the image features obtained by the camera and the real-time environmental information collected by the sensing node to comprehensively identify fire faults, which can solve the problem of high false alarm rate in fire fault identification and improve the accuracy of fire fault identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 The hardware structure diagram of a mobile terminal of a fire fault identification method based on multiple factors according to an embodiment of the present invention is

[0030] Figure 2 is a flow chart of a fire fault identification method based on multiple factors according to an embodiment of the present invention;

[0031] Figure 3 The invention is a structural block diagram of a fire fault identification device based on multiple factors according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings and in combination with the embodiments.

[0033] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0034] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 is a hardware structure block diagram of a mobile terminal of a bearing fault diagnosis method based on a diffusion model and an attention mechanism according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1Only one is shown in the figure) a processor 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) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. 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 mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0035] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as a computer program corresponding to a bearing fault diagnosis method based on a diffusion model and an attention mechanism in an embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above 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 instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the mobile 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.

[0036] 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 mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as 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.

[0037] In this embodiment, a fire fault identification method based on multiple factors is provided. Figure 2 is a flowchart of a fire fault identification method based on multiple factors according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0038] Step S201, regularly acquiring images of the area to be monitored, by shooting with a preset camera according to a preset period and a preset camera rotation route to acquire a number of images;

[0039] In this embodiment, the camera shoots according to a preset cycle and a preset camera rotation route. Each fire protection system should be equipped with a corresponding number of surveillance cameras as required to fully monitor the operation status of the fire protection system. In addition, the preset cycle can be every 5 minutes or 10 minutes. The specific time can be freely set according to the needs of the user and the fire safety level. If the user wants to obtain the operation status of the fire protection system more timely, the preset cycle needs to be set smaller, otherwise, the preset cycle needs to be set larger; if the fire safety level is higher, the preset cycle needs to be set smaller, otherwise, the preset cycle needs to be set larger. The preset camera rotates and shoots according to the preset rotation route, wherein the rotation route of the camera complements to achieve full coverage of video surveillance; multiple surveillance cameras are connected to the fire protection system video platform, and the fire protection system video platform is connected to the video picture analysis service system arranged uniformly in the control center. The control center receives the shooting pictures uploaded by multiple cameras, that is, the shooting pictures at the current moment in the fire protection system.

[0040] In this embodiment, after acquiring a plurality of images, the method further includes preprocessing the plurality of images, denoising the images using a Gaussian denoising method, and enhancing the image quality using an averaging method.

[0041] Step S202: performing feature extraction on the plurality of images, and extracting a feature value of each of the plurality of images based on a preset algorithm.

[0042] In this embodiment, each of the images is preprocessed, and then convolution processing is performed on each of the preprocessed images based on a deep convolutional network to obtain local features of each of the images; pooling processing is performed on the local features to obtain global features of each of the images; the local features and the global features are fused to obtain feature values ​​of each of the images.

[0043] In this embodiment, the characteristic value includes: the horizontal characteristic value and the vertical characteristic value of any kind of fault phenomenon, by digitally processing the video images shot in the fire protection system, and collecting the characteristic values ​​of the video fault images shot in the fire protection system in a targeted manner, wherein the characteristic value includes the horizontal characteristic value and the vertical characteristic value of a certain fault phenomenon. When the current industrial video image in the substation input into the control center in real time is obtained, since the current industrial video image contains multiple frame video images, the current frame video image input in real time among the multiple frame video images is subjected to feature extraction, and the obtained characteristic value is compared with the existing multiple fault image feature thresholds, and whether there is a fire fault in the above current frame video image is identified according to the comparison result. Among them, multiple fault image feature thresholds are respectively used as reference thresholds. If the characteristic value detected in real time is higher than a certain fault image feature threshold, it is determined that a certain substation fault occurs in the current frame video image. Otherwise, it is determined that a certain substation fault does not occur in the current frame video image.

[0044] Based on the preset algorithm, the feature value of each image is extracted. First, the convolutional neural network is used to extract the global features of the training set, that is, the features of the entire training set are extracted to obtain the global feature map dimension; then, the convolutional neural network is used to extract the local features of the training data set, that is, according to the characteristics of different fire-fighting equipment, the area around the possible fault point is used as the local feature point collection data set, and the convolutional neural network is used to extract the local features to obtain the local feature map dimension. Specifically, for global feature extraction, a feature extraction algorithm based on a multi-layer convolutional neural network is used for the images in the training set. Specifically, a single convolutional neural network is used. The convolution layer uses 5x5 as the convolution kernel, the ReLU function is used as its activation function, and the global maximum pooling is used for pooling, and finally the global features are obtained. For local feature extraction, the labeling of local foreign objects has been completed in the image annotation process. The local feature extraction still uses the feature extraction algorithm based on a multi-layer convolutional neural network. Different from the network for global feature extraction, a 3x3 convolution kernel is used in the local feature extraction process, and the ReLU function is still used as its activation function. Average pooling is used for pooling, and finally the local features of the image are obtained.

[0045] Feature fusion uses a feature conversion method based on discriminant correlation analysis (DCA) to convert the global and local features of the image, maximize the correlation of features of the same type, and minimize the redundant correlation between samples of different types. The global and local features obtained after DCA feature conversion are fused using a fully connected method to obtain the fused feature values.

[0046] Step S203, performing a first comparison on the acquired feature value of each image with a plurality of predetermined fault image feature thresholds, and performing a second comparison on the acquired feature value of each image with a feature value of an image at the next moment in the same cycle;

[0047] In this embodiment, the specific process of the first comparison is: performing a first comparison on the fused features and a plurality of predetermined fault image feature thresholds to obtain a first comparison result; if the first comparison result is that the feature value is greater than or equal to any one of the fault image feature thresholds, then the first comparison result is determined to be that there is a fire fault in the current image. At this time, it means that there are feature points in the current image that exceed the fault image feature threshold, and it can be considered that there is a fire fault in the image at this time; if the first comparison result is that the digital feature values ​​are all less than a plurality of the fault image feature thresholds, then the first comparison result is determined to be that there is no fire fault in the current image. At this time, it means that there are no feature points in the current image that exceed the fault image feature threshold, and it can be considered that there is no fire fault in the image at this time.

[0048] The specific process of the second comparison is: based on the deep convolutional network, the current image and the image at the next moment in the cycle are processed to obtain the local features of the current image and the local features of the image at the next moment in the cycle; the local features of the current image and the local features of the image at the next moment in the cycle are pooled to obtain the global features of the current image and the global features of the image at the next moment in the cycle; the local features of the current image are fused with the global features of the previous image to obtain the fused feature value of the current image, the local features of the image at the next moment in the cycle are fused with the global features of the image at the next moment in the cycle to obtain the fused feature value of the image at the next moment in the cycle, and then, a second comparison is performed between the fused feature value of the current image and the fused feature value of the image at the next moment in the cycle;

[0049] If the absolute value of the difference between the characteristic value of the current image and the characteristic value of the image at the next moment is greater than or equal to the preset threshold, it is determined that the second comparison result is that the fire fault exists in the area to be monitored; at this time, if the absolute value of the difference between the characteristic value of the current image and the characteristic value of the image at the next moment is greater than or equal to the preset threshold, it means that there is a large difference in the characteristic values ​​of the images obtained at adjacent moments in the same cycle, that is, a certain characteristic point in the area to be monitored has undergone a large change before and after, and the second comparison result is determined to be that the fire fault exists in the area to be monitored;

[0050] When the absolute value of the difference between the feature value of the current image and the feature value of the image at the next moment is less than a preset threshold, the second comparison result is determined to be that the fire fault does not exist in the area to be monitored. At this time, if the absolute value of the difference between the feature value of the current image and the feature value of the image at the next moment is less than the preset threshold, it means that the feature values ​​of the images obtained at adjacent moments in the same cycle are not much different, that is, the changes before and after a certain feature point in the area to be monitored are within a reasonable range, and the second comparison result is determined to be that the fire fault does not exist in the area to be monitored.

[0051] Step S204, performing a first fault identification based on the first comparison result and the second comparison result;

[0052] In this embodiment, the specific process of the first fault identification is: if the first comparison result and the second comparison result are both that the fire fault exists in the area to be tested, then the first fire fault identification result is that there is a fire fault; at this time, it means that the current image feature value is greater than or equal to any one of the fault image feature thresholds, and there is a large difference in the feature values ​​of the images obtained at adjacent moments in the same cycle, then it can be determined that there is a fire fault in the area to be monitored;

[0053] If at least one of the first comparison result and the second comparison result is that the fire fault does not exist in the current image, then the first fire fault identification result is pending; at this time, it means that the current image feature value is less than any of the fault image feature thresholds or the difference in feature values ​​of images obtained at adjacent moments in the same cycle is within a reasonable range, then it is impossible to determine whether there is a fire fault in the monitored area, and further judgment is needed with the help of other information.

[0054] Step S205, collecting environmental information of the area to be monitored, collecting real-time environmental information through multiple sensing nodes arranged in the area to be detected, and performing a third comparison between the real-time environmental information and the preset safety environment, and performing a second fault identification based on the third comparison result;

[0055] In this embodiment, a number of sensors are preset in the area to be monitored, and the sensors include a temperature sensor, a smoke sensor, and a carbon dioxide sensor, which are used to collect real-time environmental data. When at least one of the real-time environmental information collected by the temperature sensor, the smoke sensor, and the carbon dioxide sensor exceeds the preset safety environmental information, the second fault identification result is determined to be a fire fault, otherwise, the second fault identification result is considered to be no fire fault. In this process, if it is found that the real-time environmental information collected by any one of the temperature sensor, the smoke sensor, and the carbon dioxide sensor exceeds the preset safety environmental information, it can be said that the environment of the area to be detected has changed at this time, and it can be considered that a fire fault exists.

[0056] Step S206: Determine whether a fire-fighting device fails based on the first fault identification result and the second fault identification result.

[0057] If both the first fault identification result and the second fault identification result indicate that a firefighting fault exists, it is determined that the firefighting equipment has a fault;

[0058] If at least one of the first fault identification result and the second fault identification result is that there is no fire fault or it is pending, it is determined that the fire fault has occurred.

[0059] That is, only when both the first fault identification result and the second fault identification result simultaneously identify that a fire fault exists, the fire fault is determined to have occurred. By setting it in this way, the probability of false alarms can be well reduced, and the accuracy of fire fault identification can be improved.

[0060] In this embodiment, when it is determined that the fire-fighting equipment has failed, an alarm is immediately triggered and the alarm information is sent to the fire safety personnel via remote communication. At the same time, automated emergency response measures are taken, such as activating the fire alarm, activating the sprinkler system in the monitoring environment, opening the emergency exit, and automatically alarming.

[0061] In this embodiment, the first fault identification result is obtained through the image, and the second fault identification result is obtained through multiple sensors, and then a comprehensive judgment is made based on the first fault identification result and the second fault identification result whether there is a fire fault in the monitored area. The first fault identification result is determined based on the first comparison result of the characteristic value of the current image and the predetermined multiple fault image characteristic thresholds and the second comparison result of the characteristic value of the image at the current moment and the characteristic value of the image at the next moment in the same cycle, and the second fault identification result is based on the real-time environmental information collected by the multiple sensors. It can be seen that this embodiment combines the characteristic value of the image and the real-time environmental information collected by the multiple sensors to make a comprehensive judgment on whether a fire fault occurs in the detected area, which greatly reduces the possibility of misjudgment and makes the judgment result more accurate.

[0062] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0063] In this embodiment, a fire fault identification device based on multiple factors is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0064] Figure 3 is a structural block diagram of a fire fault identification device based on multiple factors according to an embodiment of the present invention, such as Figure 3 As shown, the device comprises:

[0065] The image acquisition module acquires a number of images by shooting with a preset camera according to a preset period and a preset camera rotation route;

[0066] A feature extraction module, extracting a feature value of each of the plurality of images based on a preset algorithm;

[0067] A comparison module, performing a first comparison between the acquired feature value of each image and a plurality of predetermined fault image feature thresholds, and performing a second comparison between the acquired feature value of each image and a feature value of an image at a next moment in the same cycle;

[0068] A first fault identification module, performing a first fault identification based on the first comparison result and the second comparison result;

[0069] A second fault identification module collects real-time environment information through a plurality of sensing nodes arranged in the area to be detected, performs a third comparison between the real-time environment information and a preset safety environment, and performs a second fault identification based on the third comparison result;

[0070] The fault determination module determines whether a fire-fighting device fails based on the first fault identification result and the second fault identification result.

[0071] In this embodiment, the fire fault identification device also includes an alarm module, which is used to immediately trigger an alarm and send the alarm information to firefighters via remote communication when it is determined that the fire equipment has a firing failure, and at the same time perform automated emergency response measures, start the sprinkler system in the monitoring environment and open the emergency exit.

[0072] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0073] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0074] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0075] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0076] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0077] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0078] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0079] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A fire fault identification method based on multiple factors, characterized in that: include: Step S201, obtaining images of the area to be monitored, by shooting with a preset camera according to a preset period and a preset camera rotation route to obtain a plurality of images; Step S202: performing feature extraction on the plurality of images, and extracting a feature value of each of the plurality of images based on a preset algorithm; Step S203, performing a first comparison on the acquired feature value of each image with a plurality of predetermined fault image feature thresholds, and performing a second comparison on the feature value of the image at the current moment with the feature value of the image at the next moment in the same cycle; Step S204: performing a first fault identification based on the first comparison result and the second comparison result; Step S205: collecting environmental information of the area to be monitored, collecting real-time environmental information through multiple sensing nodes arranged in the area to be monitored, and performing a third comparison between the real-time environmental information and preset safety environmental information, and performing a second fault identification based on the third comparison result; Step S206: determining whether a fire-fighting device fails based on the first fault identification result and the second fault identification result; The first comparing the acquired feature value of each image with a plurality of predetermined fault image feature thresholds includes: If the feature value of each of the images is greater than or equal to any of the fault image feature thresholds, determining that the first comparison result is that a fire fault exists in the area to be monitored; If the feature value of each of the images is less than the plurality of fault image feature thresholds, determining that the first comparison result is that there is no fire fault in the area to be monitored; The second comparison of the characteristic value of the image at the current moment with the characteristic value of the image at the next moment in the same period includes: Based on a deep convolutional network, convolution processing is performed on the image at the current moment and the image at the next moment in the same cycle to obtain local features of the image at the current moment and local features of the image at the next moment in the same cycle; pooling processing is performed on the local features of the image at the current moment and the local features of the image at the next moment in the same cycle respectively to obtain global features of the image at the current moment and global features of the image at the next moment in the same cycle; the local features of the image at the current moment and the global features of the image at the current moment are fused to obtain feature values ​​of the image at the current moment; the local features of the image at the next moment in the same cycle and the global features of the image at the next moment in the same cycle are fused to obtain feature values ​​of the image at the next moment in the same cycle; Perform a second comparison on the feature value of the image at the current moment and the feature value of the image at the next moment in the same period to obtain a second comparison result; If the first comparison result and the second comparison result both indicate that a fire fault exists in the area to be monitored, then the first fault identification result is that a fire fault exists in the area to be monitored; If at least one of the first comparison result and the second comparison result is that the fire fault does not exist in the monitored area, then the first fault identification result is pending; The step of performing a second comparison on the feature value of the image at the current moment and the feature value of the image at the next moment in the same period to obtain a second comparison result includes: If the absolute value of the difference between the characteristic value of the image at the current moment and the characteristic value of the image at the next moment in the same period is greater than or equal to a preset threshold, then determining that the second comparison result is that there is a fire fault in the monitored area; If the absolute value of the difference between the characteristic value of the image at the current moment and the characteristic value of the image at the next moment in the same period is less than a preset threshold, it is determined that the second comparison result is that there is no fire fault in the monitored area; The fusion includes using a feature conversion method based on discriminant correlation analysis (DCA) to perform feature conversion on the global features and local features of the image, maximizing the correlation of features of the same type, minimizing the redundant correlation between samples of different types, and using a full connection method to perform feature fusion on the global features and local features obtained after DCA feature conversion to obtain fused feature values.

2. The method according to claim 1, characterized in that The step S202 includes: Preprocess each of the images, and then perform convolution processing on each of the preprocessed images based on a deep convolutional network to obtain local features of each of the images; perform pooling processing on the local features to obtain global features of each of the images; and fuse the local features with the global features to obtain feature values ​​of each of the images.

3. The method according to claim 1, characterized in that Step S205 includes: Collecting real-time environmental information by arranging a plurality of sensing nodes in the area to be monitored, wherein the plurality of sensing nodes include: a temperature sensor, a carbon dioxide sensor and a smoke sensor; When at least one item of information collected by the temperature sensor, the carbon dioxide sensor and the two smoke sensors exceeds the preset safety environment information, the second fault identification result is determined to be that there is a fire fault in the monitored area; otherwise, the second fault identification result is determined to be that there is no fire fault in the monitored area.

4. The method according to claim 1, characterized in that: The determining whether a fire-fighting device fails based on the first fault identification result and the second fault identification includes: If both the first fault identification result and the second fault identification result indicate that a fire fault exists in the area to be monitored, it is determined that a fire fault exists in the area to be monitored; If at least one of the first fault identification result and the second fault identification result is that there is no fire fault or is to be determined, it is determined that there is no fire fault in the area to be monitored.

5. A fire fault identification device based on multiple factors, characterized in that: The method applied to any one of claims 1 to 4, wherein the fire fault identification device comprises: The image acquisition module acquires a number of images by shooting with a preset camera according to a preset period and a preset camera rotation route; A feature extraction module, extracting a feature value of each of the plurality of images based on a preset algorithm; A comparison module, performing a first comparison between the acquired feature value of each image and a plurality of predetermined fault image feature thresholds, and performing a second comparison between the acquired feature value of each image and a feature value of an image at a next moment in the same cycle; A first fault identification module, performing a first fault identification based on the first comparison result and the second comparison result; A second fault identification module collects real-time environment information through a plurality of sensing nodes arranged in the area to be detected, performs a third comparison between the real-time environment information and a preset safety environment, and performs a second fault identification based on the third comparison result; The fault determination module determines whether a fire-fighting device fails based on the first fault identification result and the second fault identification result.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program is configured to execute the method according to any one of claims 1 to 4 when executed.

7. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to perform the method according to any one of claims 1 to 4.

Images