A video data processing system based on AI big model

Abstract

The present invention relates to the technical field of video data processing systems, and in particular to a video data processing system based on an AI large model, the system comprising a fire information monitoring module, a control module, a fire determination module, and a communication module; the fire information monitoring module is used to detect indoor fire and obtain fire-related data, and transmit the fire-related data to the control module; the control module obtains a fire determination factor based on the fire-related data, and transmits the fire determination factor to the fire determination module; the fire determination module determines information on the severity of the fire based on the fire determination factor, and transmits the information on the severity of the fire to the communication module; the communication module transmits the information on the severity of the fire to a user end. The above evaluation of the severity of the fire from different directions helps to improve the accuracy of the analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of video data processing systems, and in particular to a video data processing system based on an AI big model. Background Art

[0002] The data processing system of the AI ​​large model processes the following steps: 1. Data collection and preparation, obtaining video data from cameras, etc., and preprocessing the data, including denoising, image enhancement, etc.; 2. Feature extraction, extracting features from video data, including color, motion trajectory, object boundary, texture, etc.; 3. Inter-frame association, using optical flow technology and motion estimation technology to understand the movement and change of objects; 4. Target detection and tracking, target recognition technology can identify objects in the video and mark their positions, and track the movement trajectory of objects in the video; 5. Action recognition and classification, the AI ​​model learns and recognizes different actions and classifies them, which uses key frame extraction and deep learning models.

[0003] The AI ​​big model can use image processing, computer vision and other technologies to monitor and identify the fire scene in real time, including flame detection, smoke recognition, combustible material detection, etc., so that fire-fighting measures can be taken in time.

[0004] The application document with publication number CN117576632A discloses a fire warning system and method for power grid monitoring based on a multimodal AI big model. The system includes the following steps: collecting environmental videos under the operation state of the power grid, enhancing the quality of the frame-sampling images, and encapsulating them according to standard parameters and saving them in a unified format; using the visual big model to intelligently analyze the processed images to obtain a textual description of the image content; using the textual description of the image content as input, using the text big model to perform semantic analysis on it and determine the fire risk; when there is a fire risk, using the audio big model to generate a warning voice, which is transmitted to relevant functional personnel in the form of an alarm to achieve intelligent fire warning in the power grid environment.

[0005] The existing technology for analyzing fire risks is relatively simple, which reduces the accuracy of the analysis results. Summary of the invention

[0006] The purpose of the present invention is to improve the accuracy of the analysis results. In view of the above-mentioned shortcomings, a video data processing system based on an AI large model is proposed.

[0007] The present invention adopts the following technical solution:

[0008] A video data processing system based on an AI big model, the system comprises a fire information monitoring module, a control module, a fire determination module, and a communication module.

[0009] The fire information monitoring module is used to detect the indoor fire and obtain fire-related data, and transmit the fire-related data to the control module.

[0010] The control module obtains a fire determination factor according to the fire-related data, and transmits the fire determination factor to the fire determination module.

[0011] The fire determination module determines information on the severity of the fire according to the fire determination factor, and transmits the information on the severity of the fire to the communication module.

[0012] The communication module transmits information on the severity of the fire to the user end.

[0013] Optionally, the fire information monitoring module includes a fire information setting submodule, an area storage submodule, a video data processing submodule, a smoke monitoring submodule, a temperature monitoring submodule, a sound monitoring submodule, and a gas monitoring submodule.

[0014] The fire information setting submodule is used to set the selection threshold of the AI ​​technology detection factor, the selection threshold of the sensor technology detection factor, and the ideal indoor noise value, and transmit them to the control module.

[0015] The area storage submodule is used to store the indoor area and transmit it to the control module.

[0016] The video data processing submodule is used to shoot the indoor fire and obtain the flame spread speed detected by AI technology, the maximum flame height detected by AI technology, the maximum flame temperature detected by AI technology, and the combustible index detected by AI technology, and transmit them to the control module.

[0017] The smoke monitoring submodule is used to detect indoor smoke and obtain the smoke concentration detected by the sensing technology, and transmit it to the control module.

[0018] The temperature monitoring submodule is used to detect the indoor temperature and obtain the indoor temperature detected by each temperature sensor, and transmit it to the control module.

[0019] The sound monitoring submodule is used to detect indoor noise and obtain the maximum indoor noise value detected by the sensing technology, and transmit it to the control module.

[0020] The gas monitoring submodule is used to detect indoor gas and obtain a methane reference index, a hydrogen reference index, and a carbon monoxide reference index, and transmit them to the control module.

[0021] The control module obtains the indoor combustible gas index detected by the sensing technology based on the methane reference index, the hydrogen reference index, and the carbon monoxide reference index; obtains the total number of temperature sensors based on the indoor area; obtains the average value of the indoor temperature detected by the sensing technology based on the total number of temperature sensors and the indoor temperature detected by each temperature sensor; obtains the sensing technology detection factor based on the smoke concentration detected by the sensing technology, the average value of the indoor temperature detected by the sensing technology, the maximum indoor noise value detected by the sensing technology, the ideal indoor noise value, and the indoor combustible gas index detected by the sensing technology; obtains the sensing technology evaluation factor based on the sensing technology detection factor and the selection threshold of the sensing technology detection factor; obtains the AI ​​technology detection factor based on the flame spread speed detected by the AI ​​technology, the maximum flame height detected by the AI ​​technology, the maximum flame temperature detected by the AI ​​technology, and the combustible index detected by the AI ​​technology; obtains the AI ​​technology evaluation factor based on the AI ​​technology detection factor and the selection threshold of the AI ​​technology detection factor; and obtains the fire judgment factor based on the AI ​​technology evaluation factor and the sensing technology evaluation factor.

[0022] Optionally, the video data processing submodule includes a visual processing unit and an infrared camera unit.

[0023] The visual processing unit is used to capture the indoor fire and derive the flame spread speed detected by AI technology, the maximum flame height detected by AI technology, and the combustible index detected by AI technology, and transmit them to the control module.

[0024] The infrared camera unit is used to capture the indoor fire and obtain the maximum flame temperature detected by AI technology, which is then transmitted to the control module.

[0025] Optionally, the visual processing unit includes a camera, a video preprocessor, a flame identifier, a flame motion analyzer, a flame height analyzer, a combustible material identifier, and a data outputter.

[0026] The camera is used to collect the video of the indoor fire in real time and transmit it to the video preprocessor.

[0027] The video preprocessor removes noise from the video and transmits the video to the flame identifier and the combustible material identifier.

[0028] The flame identifier utilizes visual technology to identify the flame area in the video and track the position and movement trajectory of the flame in the video.

[0029] The flame motion analyzer is used to analyze the motion trajectory of the flame and obtain the flame spread speed detected by AI technology, and transmit it to the data output device.

[0030] The flame height analyzer is used to analyze the position of the flame in the video and obtain the maximum flame height detected by AI technology, and transmit it to the data output device.

[0031] The combustible material identifier utilizes visual technology to identify combustible material signs in the video and obtains a combustible material index detected by AI technology and transmits it to a data output device.

[0032] The data output device transmits the flame spreading speed detected by the AI ​​technology, the maximum flame height detected by the AI ​​technology, and the combustible index detected by the AI ​​technology to the control module.

[0033] Optionally, the infrared camera unit includes an infrared camera, an image preprocessor, a temperature calculator, and a result analyzer.

[0034] The infrared camera captures the flame image according to the infrared spectrum information of the flame radiation, and transmits the image to the image preprocessor.

[0035] The image preprocessor performs a denoising operation on the flame image.

[0036] The temperature calculator calculates the temperature of the flame based on the pixel grayscale value of the denoised flame image and obtains the maximum flame temperature detected by the AI ​​technology, and transmits it to the result analyzer.

[0037] The result analyzer transmits the maximum flame temperature detected by the AI ​​technology to the control module.

[0038] The beneficial effects achieved by the present invention are:

[0039] With the cooperation of the fire information setting submodule, the area storage submodule, the video data processing submodule, the smoke monitoring submodule, the temperature monitoring submodule, the sound monitoring submodule and the gas monitoring submodule, fire-related data are obtained, and the fire determination factor is obtained by obtaining the fire-related data. The fire determination module determines the information of the fire severity according to the fire determination factor, and evaluates the fire severity from different directions, which helps to improve the accuracy of the analysis.

[0040] To further understand the features and technical contents of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are only for reference and description and are not intended to limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a schematic diagram of the overall structure of the present invention;

[0042] Figure 2 It is a structural schematic diagram of the fire information monitoring module in the present invention;

[0043] Figure 3 It is a structural schematic diagram of the video data processing submodule in the present invention;

[0044] Figure 4It is a structural schematic diagram of the smoke monitoring submodule in the present invention;

[0045] Figure 5 It is a schematic diagram of the overall structure of the second embodiment of the present invention. DETAILED DESCRIPTION

[0046] The following is an explanation of the embodiments of the present invention through specific embodiments. Those skilled in the art can understand the advantages and effects of the present invention from the contents disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and the details in this specification can also be modified and changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. In addition, the drawings of the present invention are only simple schematic illustrations and are not depicted according to actual dimensions. It is stated in advance. The following embodiments will further explain the relevant technical contents of the present invention in detail, but the disclosed contents are not intended to limit the scope of protection of the present invention.

[0047] Embodiment 1: This embodiment provides a video data processing system based on an AI large model, combined with Figures 1 to 4 shown.

[0048] A video data processing system based on an AI big model, the system comprises a fire information monitoring module, a control module, a fire determination module, and a communication module.

[0049] The fire information monitoring module is used to detect the indoor fire and obtain fire-related data, and transmit the fire-related data to the control module.

[0050] The control module obtains a fire determination factor according to the fire-related data, and transmits the fire determination factor to the fire determination module.

[0051] The fire determination module determines information on the severity of the fire according to the fire determination factor, and transmits the information on the severity of the fire to the communication module.

[0052] The communication module transmits information on the severity of the fire to the user end.

[0053] Specifically, when the fire determination module makes a determination, pay attention to the following matters: The values ​​are as follows: or or ,when When the fire severity is judged as general fire, When the fire severity is judged as severe fire, When the fire is judged to be a major fire, the severity of the fire is judged to be a major fire.

[0054] Optionally, the fire information monitoring module includes a fire information setting submodule, an area storage submodule, a video data processing submodule, a smoke monitoring submodule, a temperature monitoring submodule, a sound monitoring submodule, and a gas monitoring submodule.

[0055] The fire information setting submodule is used to set the selection threshold of the AI ​​technology detection factor, the selection threshold of the sensor technology detection factor, and the ideal indoor noise value, and transmit them to the control module.

[0056] The area storage submodule is used to store the indoor area and transmit it to the control module.

[0057] The video data processing submodule is used to shoot the indoor fire and obtain the flame spread speed detected by AI technology, the maximum flame height detected by AI technology, the maximum flame temperature detected by AI technology, and the combustible index detected by AI technology, and transmit them to the control module.

[0058] The smoke monitoring submodule is used to detect indoor smoke and obtain the smoke concentration detected by the sensing technology, and transmit it to the control module.

[0059] The temperature monitoring submodule is used to detect the indoor temperature and obtain the indoor temperature detected by each temperature sensor, and transmit it to the control module.

[0060] The sound monitoring submodule is used to detect indoor noise and obtain the maximum indoor noise value detected by the sensing technology, and transmit it to the control module.

[0061] The gas monitoring submodule is used to detect indoor gas and obtain a methane reference index, a hydrogen reference index, and a carbon monoxide reference index, and transmit them to the control module.

[0062] The control module obtains the indoor combustible gas index detected by the sensing technology based on the methane reference index, the hydrogen reference index, and the carbon monoxide reference index; obtains the total number of temperature sensors based on the indoor area; obtains the average value of the indoor temperature detected by the sensing technology based on the total number of temperature sensors and the indoor temperature detected by each temperature sensor; obtains the sensing technology detection factor based on the smoke concentration detected by the sensing technology, the average value of the indoor temperature detected by the sensing technology, the maximum indoor noise value detected by the sensing technology, the ideal indoor noise value, and the indoor combustible gas index detected by the sensing technology; obtains the sensing technology evaluation factor based on the sensing technology detection factor and the selection threshold of the sensing technology detection factor; obtains the AI ​​technology detection factor based on the flame spread speed detected by the AI ​​technology, the maximum flame height detected by the AI ​​technology, the maximum flame temperature detected by the AI ​​technology, and the combustible index detected by the AI ​​technology; obtains the AI ​​technology evaluation factor based on the AI ​​technology detection factor and the selection threshold of the AI ​​technology detection factor; and obtains the fire judgment factor based on the AI ​​technology evaluation factor and the sensing technology evaluation factor.

[0063] Specifically, the smoke monitoring submodule can be a smoke sensor, the temperature monitoring submodule can be a temperature sensor, the sound monitoring submodule can be a sound sensor, and the gas monitoring submodule can be a combustible gas sensor; different information can be obtained through different monitoring submodules. The greater the smoke concentration, the higher the indoor temperature, the louder the noise, and the more combustible gas, the more serious the fire.

[0064] Optionally, the video data processing submodule includes a visual processing unit and an infrared camera unit.

[0065] The visual processing unit is used to capture the indoor fire and derive the flame spread speed detected by AI technology, the maximum flame height detected by AI technology, and the combustible index detected by AI technology, and transmit them to the control module.

[0066] The infrared camera unit is used to capture the indoor fire and obtain the maximum flame temperature detected by AI technology, which is then transmitted to the control module.

[0067] Optionally, the visual processing unit includes a camera, a video preprocessor, a flame identifier, a flame motion analyzer, a flame height analyzer, a combustible material identifier, and a data outputter.

[0068] The camera is used to collect the video of the indoor fire in real time and transmit it to the video preprocessor.

[0069] The video preprocessor removes noise from the video and transmits the video to the flame identifier and the combustible material identifier.

[0070] The flame identifier utilizes visual technology to identify the flame area in the video and track the position and movement trajectory of the flame in the video.

[0071] The flame motion analyzer is used to analyze the motion trajectory of the flame and obtain the flame spread speed detected by AI technology, and transmit it to the data output device.

[0072] The flame height analyzer is used to analyze the position of the flame in the video and obtain the maximum flame height detected by AI technology, and transmit it to the data output device.

[0073] The combustible material identifier utilizes visual technology to identify combustible material signs in the video and obtains a combustible material index detected by AI technology and transmits it to a data output device.

[0074] The data output device transmits the flame spreading speed detected by the AI ​​technology, the maximum flame height detected by the AI ​​technology, and the combustible index detected by the AI ​​technology to the control module.

[0075] Specifically, AI technology is used in the identification of flame identifiers and combustible material identifiers, and the identification principles are the same. Taking the flame identifier as an example, a large amount of fire video data is first collected, covering different flame colors, shapes, sizes, and smoke density and diffusion conditions. These video data are annotated to mark the position and size of the flame and smoke for subsequent training of the algorithm model. Then, a deep learning framework is used to build a neural network model for fire identification. Common ones include convolutional neural models, which can extract useful feature information from video frames. Then, through multi-layer convolution and pooling operations, the model can learn the visual features of flames and smoke, and then optimize the model, adjusting the model parameters through different technologies. , and then cooperate the trained model with the corresponding interface, the trained model will identify each frame of video; the flame motion analyzer can use optical flow method, frame difference method, motion analysis algorithm and other technologies to analyze the movement speed and direction of the flame, and obtain the corresponding data according to the motion trajectory; Under normal circumstances, indoor fires are caused by a single ignition point. At this time, the flame identifier can identify one flame. If multiple flames appear, the required flames can be identified by setting the corresponding trained model. For example, the flame identifier can identify the flame with the largest area. The flame with the largest area refers to the flame with the largest area covered. In this embodiment, the situation of a single flame appearing indoors is studied.

[0076] Optionally, the infrared camera unit includes an infrared camera, an image preprocessor, a temperature calculator, and a result analyzer.

[0077] The infrared camera captures the flame image according to the infrared spectrum information of the flame radiation, and transmits the image to the image preprocessor.

[0078] The image preprocessor performs a denoising operation on the flame image.

[0079] The temperature calculator calculates the temperature of the flame based on the pixel grayscale value of the denoised flame image and obtains the maximum flame temperature detected by the AI ​​technology, and transmits it to the result analyzer.

[0080] The result analyzer transmits the maximum flame temperature detected by the AI ​​technology to the control module.

[0081] Optionally, when the control module calculates the fire determination factor, the following formula is satisfied:

[0082] ;

[0083] ;

[0084] ;

[0085] ;

[0086] ;

[0087] ;

[0088] ;

[0089] ;

[0090] in, is the fire determination factor, is the AI ​​technology assessment factor, Evaluation factors for sensing technologies.

[0091] The values ​​are as follows: or or ,when When the AI ​​technology evaluates the fire severity to be a general fire, When AI technology evaluates the fire severity as a major fire, It means that AI technology has assessed the severity of the fire to be a major fire. Detection factors for AI technology, and Select thresholds for detection factors for different AI techniques.

[0092] AI technology detects the speed of flame spread. The maximum flame height is detected by AI technology. The maximum flame temperature detected by AI technology, AI technology detects combustible index, The values ​​are as follows: or ,when When AI technology detects that there are no combustibles in the room, the rest are .

[0093] The values ​​are as follows: or or ,when When the sensor technology assesses the fire severity as a general fire, When the sensor technology assesses the fire severity as a major fire, When the sensor technology assesses the severity of the fire as a major fire, For the sensing technology detection factor, and Select threshold values ​​for detection factors for different sensing techniques.

[0094] The smoke concentration is detected by the sensing technology. The average indoor temperature detected by the sensing technology. The maximum indoor noise level is detected by the sensing technology. is the ideal indoor noise value, The indoor combustible gas index is detected by sensing technology.

[0095] Y is the total number of temperature sensors, For the A temperature sensor detects the indoor temperature.

[0096] is the indoor area, In square meters.

[0097] is the methane reference index, The values ​​are as follows: or ,when When the sensor technology detects that there is no methane in the room, the other cases are , is the hydrogen reference index, The values ​​are as follows: or ,when When the sensor technology detects that there is no hydrogen in the room, the other cases are , is the carbon monoxide reference index, The values ​​are as follows: or ,when When the sensor technology detects that there is no carbon monoxide in the room, the other cases are .

[0098] Specifically, "general fire" means that the fire can be controlled by the personnel of the same fire brigade in a certain area, and there is no need to mobilize other rescue forces. "Serious fire" means that the fire needs to mobilize personnel from other fire brigades, and the number of other fire brigades is limited to one, two or three. "Extreme fire" refers to other situations other than the above situations; the selection threshold of the AI ​​technology detection factor is set by technical personnel in this field; the unit of the flame spread speed detected by the AI ​​technology is meters per second; the unit of the maximum flame height detected by the AI ​​technology is meters; the unit of the maximum flame temperature detected by the AI ​​technology is degrees Celsius; the following matters need to be noted when calculating the combustible index detected by the AI ​​technology, "combustible" refers to liquid combustibles and gaseous combustibles. AI detection technology can identify the signs attached to the surface of an object to identify the corresponding liquid combustibles and gaseous combustibles; the smoke concentration detected by the sensor technology is micrograms per cubic meter; the average value of the indoor temperature detected by the sensor technology is in degrees Celsius; the maximum indoor noise value and the ideal indoor noise value detected by the sensor technology are both in decibels. In this embodiment, the ideal indoor noise value is set to 50 decibels; the indoor area refers to the total area of ​​the connected space.

[0099] The above units are only examples, and those skilled in the art may set different noise units, area units, temperature units, and smoke concentration units according to actual needs when implementing the present solution.

[0100] This embodiment solves the problem of the relatively single determination factor of the traditional system through the fire determination module, and evaluates the severity of the fire from different directions, which helps to improve the accuracy of the analysis.

[0101] Embodiment 2: This embodiment includes all the contents of Embodiment 1, and provides a video data processing system based on an AI big model, combined with Figure 5 shown.

[0102] A video data processing system based on an AI large model, wherein the fire information monitoring module also includes a humidity monitoring submodule;

[0103] The fire information setting submodule is also used to set the total number of humidity sensors and transmit it to the control module.

[0104] The humidity monitoring submodule is used to detect the indoor air humidity and obtain the indoor humidity detected by each humidity sensor and transmit it to the control module.

[0105] The control module obtains the average indoor humidity based on the total number of humidity sensors and the indoor humidity detected by each humidity sensor, obtains the flame spread reference index based on the flame spread speed detected by AI technology, and estimates the amount of water required for the fire based on the indoor area, fire judgment factor, flame spread reference index, and the average indoor humidity, and transmits the estimated amount of water required for the fire to the communication module.

[0106] The communication module transmits the estimated amount of water required for the fire to the user end.

[0107] Optionally, when the control module calculates the estimated amount of water required for a fire, the following formula is satisfied:

[0108] ;

[0109] ;

[0110] ;

[0111] in, To estimate the amount of water needed for a fire, is the ceiling function, is the indoor area, is the fire determination factor, is the flame spread reference index, is the average indoor humidity;

[0112] is meters per second;

[0113] is the total number of humidity sensors, For the A humidity sensor detects the indoor humidity.

[0114] Specifically, the unit of the estimated amount of water required for the fire is tons; the total number of humidity sensors is set by those skilled in the art.

[0115] The above units are only examples, and those skilled in the art may set different water volume units according to actual needs when implementing this solution.

[0116] The contents disclosed above are only preferred feasible embodiments of the present invention, and do not limit the protection scope of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention description and drawings are included in the protection scope of the present invention. In addition, the elements therein can be updated as technology develops.

Claims

1. A video data processing system based on AI big model, characterized in that: The system includes a fire information monitoring module, a control module, a fire determination module, and a communication module; The fire information monitoring module is used to detect the indoor fire and obtain fire-related data, and transmit the fire-related data to the control module. The fire information monitoring module includes a fire information setting submodule, an area storage submodule, a video data processing submodule, a smoke monitoring submodule, a temperature monitoring submodule, a sound monitoring submodule, and a gas monitoring submodule; The control module obtains a fire determination factor based on the fire-related data and transmits the fire determination factor to the fire determination module; The fire determination module determines information on the severity of the fire according to the fire determination factor, and transmits the information on the severity of the fire to the communication module; The communication module transmits information about the severity of the fire to the user end; The control module calculates a fire determination factor based on an indoor combustible gas index detected by the sensing technology, an average value of an indoor temperature detected by the sensing technology, a sensing technology detection factor, a sensing technology evaluation factor, an AI technology detection factor, and an AI technology evaluation factor; When the control module calculates the fire determination factor, the following formula is satisfied: ； in, is the fire determination factor, is the AI ​​technology assessment factor, Evaluate factors for sensing technology; The fire information setting submodule is used to set the selection threshold of the AI ​​technology detection factor, the selection threshold of the sensor technology detection factor, and the ideal indoor noise value, and transmit them to the control module; The area storage submodule is used to store the indoor area and transmit it to the control module; The video data processing submodule is used to shoot the indoor fire and obtain the flame spread speed detected by AI technology, the maximum flame height detected by AI technology, the maximum flame temperature detected by AI technology, and the combustible index detected by AI technology, and transmit them to the control module; The smoke monitoring submodule is used to detect indoor smoke and obtain the smoke concentration detected by the sensing technology and transmit it to the control module; The temperature monitoring submodule is used to detect the indoor temperature and obtain the indoor temperature detected by each temperature sensor, and transmit it to the control module; The sound monitoring submodule is used to detect indoor noise and obtain the maximum indoor noise value detected by the sensing technology, and transmit it to the control module; The gas monitoring submodule is used to detect indoor gas and obtain a methane reference index, a hydrogen reference index, and a carbon monoxide reference index, and transmit them to the control module; The control module obtains the indoor combustible gas index detected by the sensing technology according to the methane reference index, the hydrogen reference index, and the carbon monoxide reference index, obtains the total number of temperature sensors according to the indoor area, obtains the average value of the indoor temperature detected by the sensing technology according to the total number of temperature sensors and the indoor temperature detected by each temperature sensor, obtains the sensing technology detection factor according to the smoke concentration detected by the sensing technology, the average value of the indoor temperature detected by the sensing technology, the maximum indoor noise value detected by the sensing technology, the ideal indoor noise value, and the indoor combustible gas index detected by the sensing technology, obtains the sensing technology evaluation factor according to the sensing technology detection factor and the selection threshold of the sensing technology detection factor, obtains the AI ​​technology detection factor according to the flame spread speed detected by the AI ​​technology, the maximum flame height detected by the AI ​​technology, the maximum flame temperature detected by the AI ​​technology, and the combustible index detected by the AI ​​technology, obtains the AI ​​technology detection factor according to the AI ​​technology detection factor and the selection threshold of the AI ​​technology detection factor, and obtains the fire determination factor according to the AI ​​technology evaluation factor and the sensing technology evaluation factor; The video data processing submodule includes a visual processing unit and an infrared camera unit; The visual processing unit is used to photograph the indoor fire and obtain the flame spread speed detected by AI technology, the maximum flame height detected by AI technology, and the combustible index detected by AI technology, and transmit them to the control module; The infrared camera unit is used to photograph the indoor fire and obtain the maximum flame temperature detected by AI technology, and transmit it to the control module; The visual processing unit includes a camera, a video preprocessor, a flame identifier, a flame motion analyzer, a flame height analyzer, a combustible material identifier, and a data output device.

2. A video data processing system based on an AI large model as claimed in claim 1, characterized in that: The camera is used to collect the video of the indoor fire in real time and transmit it to the video preprocessor; The video preprocessor removes noise from the video and transmits it to the flame identifier and the combustible material identifier; The flame identifier uses visual technology to identify the flame area in the video and track the position and movement trajectory of the flame in the video; The flame motion analyzer is used to analyze the motion trajectory of the flame and obtain the flame spread speed detected by AI technology, and transmit it to the data output device; The flame height analyzer is used to analyze the position of the flame in the video and obtain the maximum flame height detected by the AI ​​technology, and transmit it to the data output device; The combustible material identifier uses visual technology to identify combustible material signs in the video and obtains the combustible material index detected by AI technology and transmits it to the data output device; The data output device transmits the flame spreading speed detected by the AI ​​technology, the maximum flame height detected by the AI ​​technology, and the combustible index detected by the AI ​​technology to the control module.

3. A video data processing system based on AI big model as claimed in claim 2, characterized in that: The infrared camera unit includes an infrared camera, an image preprocessor, a temperature calculator, and a result analyzer; The infrared camera captures the flame image according to the infrared spectrum information of the flame radiation and transmits it to the image preprocessor; The image preprocessor performs a denoising operation on the flame image; The temperature calculator calculates the temperature of the flame based on the pixel grayscale value of the denoised flame image and obtains the maximum flame temperature detected by the AI ​​technology, and transmits it to the result analyzer; The result analyzer transmits the maximum flame temperature detected by the AI ​​technology to the control module.

Images