Door and window anti-explosion monitoring and early warning method and system

Through gas sensors, smoke sensors and image recognition technology, intelligent monitoring and early warning of door and window environments is achieved, which solves the problem of difficulty in dealing with high-risk places with traditional monitoring methods and improves safety and monitoring efficiency.

CN119068623APending Publication Date: 2024-12-03SHANGHAI LINTAO INTELLIGENT DOOR & WINDOW CURTAIN WALL CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411241803.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-05
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

Traditional door and window monitoring methods lack intelligence, making it difficult to deal with potential safety risks in a timely and effective manner, especially in high-risk places, and it is difficult to achieve real-time monitoring and early warning of gases and smoke.

Method used

Gas sensor group and smoke sensor are used to monitor the gas and smoke concentration in the door and window environment in real time, and combine high-definition cameras and image recognition technology to determine whether there are fire elements in the image through neural network models, achieving intelligent monitoring and early warning of doors and windows explosion-proof.

Benefits of technology

It realizes multi-dimensional and comprehensive monitoring of door and window environments, improves the accuracy and reliability of early warnings, promptly detects and warns of potential explosion or fire risks, ensures personnel and property safety, reduces labor costs, and improves monitoring efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119068623A_ABST
    Figure CN119068623A_ABST
Patent Text Reader

Abstract

The invention provides a door and window anti-explosion monitoring and early warning method and system. The door and window anti-explosion monitoring and early warning method comprises the steps that gas concentration information and smoke concentration information corresponding to the environment where a door and a window are located are monitored in real time through a gas sensor set and a smoke sensor, and whether anti-explosion alarming is conducted or not is judged according to the gas concentration information and the smoke concentration information corresponding to the environment where the door and the window are located; image data of a door and window area range are collected in real time through an image collection device, and whether anti-explosion alarm is carried out or not is judged through image information. The system comprises modules corresponding to the steps of the method.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention provides a door and window explosion-proof monitoring and early warning method and system, belonging to the technical field of explosion-proof control. Background Art

[0002] With the acceleration of urbanization and the continuous improvement of building safety standards, doors and windows, as an important part of buildings, are becoming more and more concerned about their safety and stability. Especially in some high-risk places, such as chemical plants, oil and gas stations, warehouses, etc., the explosion-proof performance of doors and windows is directly related to the safety of personnel and property. However, traditional door and window monitoring methods are often limited to simple physical protection, lacking intelligent monitoring and early warning functions, and it is difficult to respond to potential safety risks in a timely and effective manner. Summary of the invention

[0003] The present invention provides a door and window explosion-proof monitoring and early warning method and system to solve the above-mentioned technical problems existing in the prior art. The technical solutions adopted are as follows:

[0004] The door and window explosion-proof monitoring and early warning method comprises:

[0005] The gas concentration information and smoke concentration information corresponding to the environment in which the doors and windows are located are monitored in real time by the gas sensor group and the smoke sensor, and it is determined whether to issue an explosion-proof alarm according to the gas concentration information and smoke concentration information corresponding to the environment in which the doors and windows are located;

[0006] The image data of the door and window area is collected in real time by the image collection device, and the image information is used to determine whether to issue an explosion-proof alarm.

[0007] Furthermore, the gas concentration information and smoke concentration information corresponding to the environment in which the doors and windows are located are monitored in real time by the gas sensor group and the smoke sensor, and whether to perform an explosion-proof alarm is determined according to the gas concentration information and smoke concentration information corresponding to the environment in which the doors and windows are located, including:

[0008] The gas sensor group is used to monitor the gas concentration information corresponding to the environment in which the doors and windows are located in real time; wherein the gas sensor group includes a methane sensor, a carbon monoxide sensor, an ethane sensor and a propane sensor;

[0009] When the gas concentration information corresponding to the environment in which the doors and windows are located exceeds the preset gas concentration threshold, an explosion-proof alarm is issued;

[0010] The smoke concentration information is used to monitor the smoke concentration information corresponding to the environment in which the doors and windows are located in real time;

[0011] When the smoke concentration information corresponding to the environment in which the doors and windows are located exceeds a preset smoke concentration threshold, an explosion-proof alarm is issued.

[0012] Furthermore, the image data of the door and window area is collected in real time by an image acquisition device, and it is determined whether to perform an explosion-proof alarm based on the image information, including:

[0013] The image acquisition device acquires video image data of the door and window area in real time; wherein the image acquisition data is acquired using a high-definition camera;

[0014] Performing frame processing on the video image data to obtain frame image data corresponding to the video image data;

[0015] Performing image enhancement processing on the frame image data to obtain frame image data after image enhancement processing;

[0016] Inputting the frame image data after the image enhancement processing into a neural network model for image recognition, and judging whether there is a fire element in the frame image data by the neural network model;

[0017] Determine whether to issue an explosion-proof alarm based on the frame image data of the fire element.

[0018] Further, performing image enhancement processing on the frame image data to obtain the frame image data after image enhancement processing includes:

[0019] Performing grayscale processing on the frame image data to obtain a grayscale value corresponding to each pixel point contained in the frame image data;

[0020] The image segmentation coefficient is obtained according to the grayscale value corresponding to each pixel point contained in the frame image data, wherein the image segmentation coefficient is obtained by the following formula:

[0021]

[0022] Where, f represents the image segmentation coefficient; n represents the number of pixels contained in the frame image data; H i Indicates the gray value corresponding to the i-th pixel; H i-1 Indicates the gray value corresponding to the i-1th pixel; H max Indicates the maximum grayscale value of the pixel points contained in the frame image data; H z Indicates the central grayscale value of the pixel contained in the frame image data;

[0023] Extracting an initial image block size corresponding to the frame image data; wherein the initial image block size is 5×5;

[0024] The image block size corresponding to the current frame image data is obtained by using the image segmentation coefficient combined with the initial image block size; wherein the image block size corresponding to the current frame image data is obtained by the following formula:

[0025]

[0026] Wherein, N represents the number of pixels in the horizontal and vertical directions corresponding to the image block size corresponding to the current frame image data, and N is rounded up, and the minimum value of N is 3; f represents the image segmentation coefficient; N 0 Indicates the number of pixels in the horizontal and vertical directions corresponding to the size of the initial image block;

[0027] Segmenting the frame image data according to the image block size corresponding to the current frame image data to obtain a plurality of image blocks corresponding to the current frame image data;

[0028] Each image block corresponding to the current frame image data is traversed, and each image block corresponding to the current frame image data is subjected to image enhancement processing to obtain the frame image data after image enhancement processing; wherein the image enhancement processing includes noise suppression processing, contrast enhancement processing and sharpening processing.

[0029] Further, judging whether to perform explosion-proof alarm according to the frame image data of the fire element includes:

[0030] Extracting the frame image data with the fire element as the target frame image data;

[0031] Sorting the target frame image data in order from front to back of the frame time to obtain an image data set corresponding to the target frame image data;

[0032] Extract the number of positions corresponding to the fire light elements contained in each target frame image data and the number of pixel points contained in the fire light elements corresponding to each position;

[0033] The fire light level evaluation coefficient corresponding to each target frame image data is obtained according to the number of positions corresponding to the fire light elements contained in each target frame image data and the number of pixels contained in the fire light elements corresponding to each position, wherein the fire light level evaluation coefficient is obtained by the following formula:

[0034]

[0035] Where E represents the fire degree evaluation coefficient; m represents the number of target frame image data; M i Indicates the number of positions corresponding to the fire element corresponding to the i-th target frame image data; M i-1 Indicates the number of positions corresponding to the fire element corresponding to the i-1th target frame image data; B i represents the number of pixels contained in the position fire element corresponding to the jth fire element in the i-th target frame image data; B irepresents the number of pixels contained in the position fire element corresponding to the j-1th fire element in the i-th target frame image data; B m Indicates the number of all pixels corresponding to the fire element of the most recent target frame image data corresponding to the frame time in the target frame image data; B c Indicates the number of all pixel points corresponding to the fire element of the first target frame image data corresponding to the frame time in the target frame image data;

[0036] Comparing the fire light level evaluation coefficient corresponding to each target frame image data with a preset fire light level evaluation coefficient threshold in turn;

[0037] When the fire light level evaluation coefficient corresponding to the target frame image data exceeds a preset fire light level evaluation coefficient threshold, an explosion-proof alarm is issued;

[0038] When the fire light level evaluation coefficients corresponding to the target frame image data that have completed the comparison do not exceed the preset fire light level evaluation coefficient threshold, the fire light level evaluation coefficient corresponding to each target frame image data is extracted, and the fire light level evaluation coefficient corresponding to each target frame image data is used to obtain a comprehensive evaluation coefficient, wherein the comprehensive evaluation coefficient is obtained by the following formula:

[0039]

[0040] Among them, P represents the comprehensive evaluation coefficient; E m E represents the fire degree evaluation coefficient of the most recent target frame image data corresponding to the frame time in the target frame image data; y represents the preset fire light evaluation coefficient threshold; k represents the number of target frame image data that have been compared; ΔT si represents the time length between the frame time corresponding to the i-th target frame image data and the i+1-th frame time; E i represents the fire degree evaluation coefficient corresponding to the i-th target frame image data; E mc Indicates the fire light evaluation coefficient corresponding to the previous target frame image data of the most recent target frame image data; E c Indicates the fire degree evaluation coefficient corresponding to the first target frame image data;

[0041] When the comprehensive evaluation coefficient exceeds a preset comprehensive evaluation coefficient threshold, an explosion-proof alarm is issued.

[0042] Door and window explosion-proof monitoring and early warning system, the door and window explosion-proof monitoring and early warning system includes:

[0043] The first explosion-proof alarm module is used to monitor the gas concentration information and smoke concentration information corresponding to the environment where the doors and windows are located in real time through the gas sensor group and the smoke sensor, and determine whether to perform an explosion-proof alarm according to the gas concentration information and smoke concentration information corresponding to the environment where the doors and windows are located;

[0044] The second explosion-proof alarm module is used to collect image data of the door and window area in real time through an image acquisition device, and determine whether to issue an explosion-proof alarm based on the image information.

[0045] Furthermore, the first explosion-proof alarm module includes:

[0046] A gas concentration information acquisition module, used to monitor the gas concentration information corresponding to the environment of the doors and windows in real time through the gas sensor group; wherein the gas sensor group includes a methane sensor, a carbon monoxide sensor, an ethane sensor and a propane sensor;

[0047] A first explosion-proof alarm execution module is used to issue an explosion-proof alarm when the gas concentration information corresponding to the environment in which the doors and windows are located exceeds a preset gas concentration threshold;

[0048] A smoke concentration information acquisition module is used to monitor the smoke concentration information corresponding to the environment where the doors and windows are located in real time through the smoke concentration information;

[0049] The second explosion-proof alarm execution module is used to issue an explosion-proof alarm when the smoke concentration information corresponding to the environment in which the doors and windows are located exceeds a preset smoke concentration threshold.

[0050] Furthermore, the second explosion-proof alarm module includes:

[0051] An image acquisition control module is used for the image acquisition device to acquire video image data of the door and window area in real time; wherein the image acquisition data is acquired using a high-definition camera;

[0052] A frame processing module, used for performing frame processing on the video image data to obtain frame image data corresponding to the video image data;

[0053] An image enhancement processing module is used to perform image enhancement processing on the frame image data to obtain frame image data after image enhancement processing;

[0054] A fire element judgment module is used to input the frame image data after the image enhancement processing into a neural network model for image recognition, and judge whether there is a fire element in the frame image data through the neural network model;

[0055] The explosion-proof alarm execution module is used to determine whether to issue an explosion-proof alarm based on the frame image data of the fire element.

[0056] Furthermore, the image enhancement processing module includes:

[0057] A grayscale processing module, used to perform grayscale processing on the frame image data to obtain a grayscale value corresponding to each pixel point contained in the frame image data;

[0058] The image segmentation coefficient acquisition module is used to obtain the image segmentation coefficient according to the gray value corresponding to each pixel point contained in the frame image data, wherein the image segmentation coefficient is obtained by the following formula:

[0059]

[0060] Where, f represents the image segmentation coefficient; n represents the number of pixels contained in the frame image data; H i Indicates the gray value corresponding to the i-th pixel; H i-1 Indicates the gray value corresponding to the i-1th pixel; H max Indicates the maximum grayscale value of the pixel points contained in the frame image data; H z Indicates the central grayscale value of the pixel contained in the frame image data;

[0061] An initial image block size acquisition module, used to extract the initial image block size corresponding to the frame image data; wherein the initial image block size is 5×5;

[0062] The image block size acquisition module is used to obtain the image block size corresponding to the current frame image data by using the image segmentation coefficient combined with the initial image block size; wherein the image block size corresponding to the current frame image data is obtained by the following formula:

[0063]

[0064] Wherein, N represents the number of pixels in the horizontal and vertical directions corresponding to the image block size corresponding to the current frame image data, and N is rounded up, and the minimum value of N is 3; f represents the image segmentation coefficient; N 0 Indicates the number of pixels in the horizontal and vertical directions corresponding to the size of the initial image block;

[0065] An image segmentation module, used to segment the frame image data according to the image block size corresponding to the current frame image data, and obtain a plurality of image blocks corresponding to the current frame image data;

[0066] The block image enhancement module is used to traverse each image block corresponding to the current frame image data, and perform image enhancement processing on each image block corresponding to the current frame image data to obtain the frame image data after image enhancement processing; wherein the image enhancement processing includes noise suppression processing, contrast enhancement processing and sharpening processing.

[0067] Furthermore, the explosion-proof alarm execution module includes:

[0068] A target frame image data determination module is used to extract frame image data with fire light elements as target frame image data;

[0069] An image sorting module is used to sort the target frame image data in order from front to back in frame time, and obtain an image data set corresponding to the target frame image data;

[0070] An information extraction module is used to extract the number of positions corresponding to the fire light elements contained in each target frame image data and the number of pixel points contained in the fire light elements corresponding to each position;

[0071] The fire light degree evaluation coefficient acquisition module is used to obtain the fire light degree evaluation coefficient corresponding to each target frame image data according to the number of positions corresponding to the fire light elements contained in each target frame image data and the number of pixels contained in the fire light elements corresponding to each position, wherein the fire light degree evaluation coefficient is obtained by the following formula:

[0072]

[0073] Where E represents the fire degree evaluation coefficient; m represents the number of target frame image data; M i Indicates the number of positions corresponding to the fire element corresponding to the i-th target frame image data; M i-1 Indicates the number of positions corresponding to the fire element corresponding to the i-1th target frame image data; B i represents the number of pixels contained in the position fire element corresponding to the jth fire element in the i-th target frame image data; B i represents the number of pixels contained in the position fire element corresponding to the j-1th fire element in the i-th target frame image data; B m Indicates the number of all pixels corresponding to the fire element of the most recent target frame image data corresponding to the frame time in the target frame image data; B c Indicates the number of all pixel points corresponding to the fire element of the first target frame image data corresponding to the frame time in the target frame image data;

[0074] A comparison module, used for comparing the fire light level evaluation coefficient corresponding to each target frame image data with a preset fire light level evaluation coefficient threshold in turn;

[0075] A first alarm execution module, configured to issue an explosion-proof alarm when the fire light level evaluation coefficient corresponding to the target frame image data exceeds a preset fire light level evaluation coefficient threshold;

[0076] The comprehensive evaluation coefficient acquisition module is used to extract the fire light degree evaluation coefficient corresponding to each target frame image data when the fire light degree evaluation coefficients corresponding to the target frame image data that have completed the comparison do not exceed the preset fire light degree evaluation coefficient threshold, and use the fire light degree evaluation coefficient corresponding to each target frame image data to obtain the comprehensive evaluation coefficient, wherein the comprehensive evaluation coefficient is obtained by the following formula:

[0077]

[0078] Among them, P represents the comprehensive evaluation coefficient; E m E represents the fire degree evaluation coefficient of the most recent target frame image data corresponding to the frame time in the target frame image data; y represents the preset fire light evaluation coefficient threshold; k represents the number of target frame image data that have been compared; ΔT si represents the time length between the frame time corresponding to the i-th target frame image data and the i+1-th frame time; E i represents the fire degree evaluation coefficient corresponding to the i-th target frame image data; E mc Indicates the fire light evaluation coefficient corresponding to the previous target frame image data of the most recent target frame image data; E c Indicates the fire degree evaluation coefficient corresponding to the first target frame image data;

[0079] The second alarm execution module is used to issue an explosion-proof alarm when the comprehensive evaluation coefficient exceeds a preset comprehensive evaluation coefficient threshold.

[0080] Beneficial effects of the present invention:

[0081] The door and window explosion-proof monitoring and early warning method and system proposed by the present invention can timely detect and warn potential explosion or fire risks by real-time monitoring of gas concentration, smoke concentration and image information in the environment, thereby effectively avoiding the occurrence of safety accidents and ensuring the safety of personnel and property. Combined with a variety of monitoring means, the method can monitor the environment in which the doors and windows are located from multiple dimensions in an all-round way. This multi-source information fusion method can significantly improve the accuracy and reliability of early warning. Once the system detects an abnormal situation and triggers an alarm, relevant departments and personnel can respond quickly and take measures to prevent the situation from further deteriorating. This rapid response mechanism is of great significance for reducing losses. The method integrates advanced technologies such as the Internet of Things and image recognition to realize intelligent management of door and window explosion-proof monitoring. This not only improves monitoring efficiency, but also reduces labor costs. By providing timely and accurate early warning information, the method can enhance users' confidence and satisfaction with home safety. At the same time, intelligent monitoring means also provide users with a more convenient and efficient management method. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 is a flow chart of the method of the present invention;

[0083] Figure 2 The system block diagram of the system of the present invention. DETAILED DESCRIPTION

[0084] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0085] The embodiment of the present invention proposes a door and window explosion-proof monitoring and early warning method, such as Figure 1 As shown, the door and window explosion-proof monitoring and early warning method includes:

[0086] S1. Real-time monitoring of the gas concentration information and smoke concentration information corresponding to the environment in which the doors and windows are located by means of a gas sensor group and a smoke sensor, and judging whether to perform an explosion-proof alarm according to the gas concentration information and smoke concentration information corresponding to the environment in which the doors and windows are located;

[0087] S2. The image data of the door and window area is collected in real time by an image collection device, and it is determined whether to issue an explosion-proof alarm based on the image information.

[0088] The working principle of the above technical solution is: through the gas sensor group and smoke sensor deployed near the doors and windows, the gas concentration information and smoke concentration information in the environment are monitored in real time. These sensors can accurately sense the concentration changes of harmful gases (such as combustible gases, toxic gases, etc.) and smoke particles in the air. The gas concentration and smoke concentration monitored by the system in real time are compared with the preset safety threshold. If any monitoring value exceeds the threshold, the system determines that there is a safety hazard in the current environment, which may cause an explosion or fire, and then trigger an explosion-proof alarm.

[0089] Image data of the door and window area is collected in real time using an image acquisition device (such as a camera). These image data contain visual information of the doors and windows and their surroundings, and can reflect the actual situation on site. The system intelligently analyzes the collected images and identifies possible abnormal conditions, such as flames. If the image analysis results show a potential risk of explosion or fire, the system will also trigger an explosion-proof alarm.

[0090] The effect of the above technical solution is: by real-time monitoring of the gas concentration, smoke concentration and image information in the environment, the method can timely discover and warn of potential explosion or fire risks, thereby effectively avoiding the occurrence of safety accidents and ensuring the safety of personnel and property. Combined with a variety of monitoring methods, the method can monitor the environment of doors and windows from multiple dimensions. This multi-source information fusion method can significantly improve the accuracy and reliability of early warning. Once the system detects an abnormal situation and triggers an alarm, relevant departments and personnel can respond quickly and take measures to prevent the situation from further deteriorating. This rapid response mechanism is of great significance for reducing losses. The method integrates advanced technologies such as the Internet of Things and image recognition to realize intelligent management of explosion-proof monitoring of doors and windows. This not only improves monitoring efficiency, but also reduces labor costs. By providing timely and accurate early warning information, the method can enhance users' confidence and satisfaction with home safety. At the same time, intelligent monitoring methods also provide users with a more convenient and efficient management method.

[0091] In one embodiment of the present invention, the gas concentration information and smoke concentration information corresponding to the environment in which the doors and windows are located are monitored in real time by a gas sensor group and a smoke sensor, and whether to perform an explosion-proof alarm is determined according to the gas concentration information and smoke concentration information corresponding to the environment in which the doors and windows are located, including:

[0092] S101, monitoring the gas concentration information corresponding to the environment of the doors and windows in real time through the gas sensor group; wherein the gas sensor group includes hydrocarbon gas sensors such as methane sensors, carbon monoxide sensors, ethane sensors and propane sensors;

[0093] S102, when the gas concentration information corresponding to the environment in which the doors and windows are located exceeds a preset gas concentration threshold, an explosion-proof alarm is issued;

[0094] S103, real-time monitoring of smoke concentration information corresponding to the environment in which the doors and windows are located through the smoke concentration information;

[0095] S104: When the smoke concentration information corresponding to the environment in which the doors and windows are located exceeds a preset smoke concentration threshold, an explosion-proof alarm is issued.

[0096] The working principle of the above technical solution is: the system uses a gas sensor group (including hydrocarbon gas sensors such as methane sensors, carbon monoxide sensors, ethane sensors and propane sensors) to monitor the gas concentration information in the environment where the doors and windows are located in real time. These sensors can accurately sense the concentration changes of specific gases in the air and provide data support for subsequent judgments.

[0097] The system compares the real-time monitored gas concentration information with the preset gas concentration thresholds. These thresholds are set according to relevant safety standards and experience to determine whether the gas concentration in the current environment is within a safe range. If any gas concentration exceeds its corresponding threshold, the system determines that there is a potential explosion risk and immediately triggers an explosion-proof alarm. While monitoring the gas concentration, the system also monitors the smoke concentration information in the environment where the doors and windows are located in real time through smoke sensors. Smoke sensors can sense smoke particles in the air and convert them into electrical signals for processing.

[0098] The system compares the real-time monitored smoke concentration information with the preset smoke concentration threshold. Similar to the gas concentration threshold, the smoke concentration threshold is also set based on safety standards and experience. If the smoke concentration exceeds its corresponding threshold, the system determines that there is a potential fire risk and also triggers the explosion-proof alarm.

[0099] The effect of the above technical solution is: by real-time monitoring of the gas concentration and smoke concentration in the environment, the method can timely detect and warn of potential explosion or fire risks, thereby effectively avoiding the occurrence of safety accidents and ensuring the safety of personnel and property. The use of multiple sensors to monitor different parameters in the environment and judge them in combination with preset thresholds can significantly improve the accuracy and reliability of the early warning. Once the system detects an abnormal situation and triggers an alarm, the relevant departments and personnel can respond quickly and take measures to prevent the situation from further deteriorating. This rapid response mechanism is of great significance for reducing losses. The method integrates the Internet of Things technology to realize intelligent monitoring and management of the environment in which doors and windows are located. This not only improves monitoring efficiency, but also reduces labor costs. By adjusting the preset threshold, the system can be flexibly configured according to different application scenarios and security requirements to adapt to different monitoring needs. By providing timely and accurate early warning information, the method can enhance users' confidence and satisfaction with home safety. At the same time, intelligent monitoring methods also provide users with a more convenient and efficient management method.

[0100] In one embodiment of the present invention, image data of a door and window area is collected in real time by an image collection device, and whether to issue an explosion-proof alarm is determined based on the image information, including:

[0101] S201, an image acquisition device acquires video image data of a door and window area in real time; wherein the image acquisition data is acquired using a high-definition camera;

[0102] S202, performing frame processing on the video image data to obtain frame image data corresponding to the video image data;

[0103] S203, performing image enhancement processing on the frame image data to obtain frame image data after image enhancement processing;

[0104] S204, inputting the frame image data after the image enhancement processing into a neural network model for image recognition, and determining whether there is a fire element in the frame image data through the neural network model;

[0105] S205: Determine whether to issue an explosion-proof alarm based on the frame image data of the fire element.

[0106] Specifically, the neural network model structure for image recognition is as follows:

[0107] Input layer: The input layer is responsible for receiving the frame image data after image enhancement. Since the frame image data is two-dimensional (height x width), it is usually necessary to convert it into a one-dimensional vector (for example, by flattening the pixel values) or use a specific data format (such as multiple one-dimensional vectors after RGB channels are separated) as the input of the neural network.

[0108] The size of the input layer (i.e., the number of neurons) depends on the dimension and format of the frame image data. For example, if the frame image size is 640x480 pixels and is an RGB three-channel image, the input layer may need to receive input from 6404803=921600 neurons (if flattened directly), or three input vectors of size 640*480=307200 (if separated by channels).

[0109] Hidden layer: Hidden layer is the key part of neural network for feature extraction and learning. The number of hidden layers and the number of neurons in each layer can be determined according to the complexity of the specific task and the required learning ability.

[0110] In image recognition tasks, common hidden layer types include convolutional layers, pooling layers, and fully connected layers. The convolutional layer extracts local features in the image through convolution operations; the pooling layer reduces the dimension of the feature map through pooling operations, reduces the amount of calculation and extracts the main features; the fully connected layer maps the learned feature representation to the sample label space, playing the role of a classifier.

[0111] The specific configuration of the hidden layer (such as the number of layers, the number of neurons in each layer, activation function, etc.) needs to be determined through experiments and parameter adjustment to achieve the best recognition effect.

[0112] Output layer: The output layer is responsible for outputting the prediction results of the neural network. In image recognition tasks, especially in the task of determining whether there is a fire element in the frame image data, the output layer usually contains one or more neurons.

[0113] If the task is a binary classification problem (i.e., there is a fire element or there is no fire element), the output layer contains one neuron and uses the sigmoid activation function to output a probability value between 0 and 1, indicating the possibility of the presence of a fire element.

[0114] The working principle of the above technical solution is: use a high-definition camera as an image acquisition device to capture video image data of the door and window area in real time. The high-definition camera can provide clear and delicate images, which is conducive to the accuracy and efficiency of subsequent image processing. Split the continuous video image data into separate frame image data. Frame is the basic unit of video, and each frame represents a static image. Through frame processing, the video data can be converted into a series of static image data for subsequent processing. Image enhancement processing is performed on the frame image data to improve the visual effect of the image and the accuracy of feature extraction. Image enhancement processing may include operations such as denoising, sharpening, and contrast adjustment, aiming to make the fire element in the image more prominent and easy to identify. The frame image data after image enhancement processing is input into the neural network model for image recognition. The neural network model is trained to identify specific elements in the image, such as fire. The model analyzes the input frame image and determines whether there is a fire element in it. According to the judgment result of the neural network model, if there is a fire element in the frame image data, the system believes that there is a potential fire risk in the current environment and triggers an explosion-proof alarm. The alarm signal can be carried out by sound, light signal or sending alarm information to designated equipment.

[0115] The effect of the above technical solution is: through high-definition cameras and image processing technology, image data of the door and window area can be captured in real time and clearly presented. Combined with the powerful recognition ability of the neural network model, it can accurately determine whether there are fire elements in the image, thereby improving the accuracy of fire detection. Once the fire element is detected and the alarm is triggered, the relevant departments and personnel can respond quickly and take measures to prevent the further spread of the fire. This rapid response mechanism is of great significance for reducing losses. Using the neural network model for image recognition, intelligent monitoring and management of the door and window area is realized. This not only improves the monitoring efficiency, but also reduces the labor cost. Through real-time monitoring and early warning, the method can timely detect and deal with potential fire risks, thereby enhancing the safety of people and property. Intelligent monitoring means provide users with a more convenient and efficient management method. At the same time, timely early warning information can also enhance users' confidence and satisfaction with home safety.

[0116] In one embodiment of the present invention, performing image enhancement processing on the frame image data to obtain the frame image data after image enhancement processing includes:

[0117] S2031, performing grayscale processing on the frame image data to obtain a grayscale value corresponding to each pixel point contained in the frame image data;

[0118] S2032. Obtain an image segmentation coefficient according to the grayscale value corresponding to each pixel point included in the frame image data, wherein the image segmentation coefficient is obtained by the following formula:

[0119]

[0120] Where, f represents the image segmentation coefficient; n represents the number of pixels contained in the frame image data; H i Indicates the gray value corresponding to the i-th pixel; H i-1 Indicates the gray value corresponding to the i-1th pixel; H max Indicates the maximum grayscale value of the pixel points contained in the frame image data; H z Indicates the central grayscale value of the pixel contained in the frame image data;

[0121] S2033, extracting an initial image block size corresponding to the frame image data; wherein the initial image block size is 5×5;

[0122] S2034, using the image segmentation coefficient combined with the initial image block size to obtain the image block size corresponding to the current frame image data; wherein the image block size corresponding to the current frame image data is obtained by the following formula:

[0123]

[0124] Wherein, N represents the number of pixels in the horizontal and vertical directions corresponding to the image block size corresponding to the current frame image data, and N is rounded up, and the minimum value of N is 3; f represents the image segmentation coefficient; N 0 Indicates the number of pixels in the horizontal and vertical directions corresponding to the size of the initial image block;

[0125] S2035, dividing the frame image data according to the image block size corresponding to the current frame image data, and obtaining a plurality of image blocks corresponding to the current frame image data;

[0126] S2036. Traverse each image block corresponding to the current frame image data, and perform image enhancement processing on each image block corresponding to the current frame image data to obtain the frame image data after image enhancement processing; wherein the image enhancement processing includes noise suppression processing, contrast enhancement processing and sharpening processing.

[0127] The working principle of the above technical solution is: convert the frame image data from color to grayscale image. Grayscale processing simplifies the image information, and each pixel is represented by a grayscale value, which is convenient for subsequent processing. According to the grayscale value of each pixel in the grayscale image, the image segmentation coefficient (f) is calculated by a specific formula. This coefficient takes into account the change of the grayscale value of the pixel, the maximum grayscale value (Hmax) and the central grayscale value (Hz), which is used to adjust the image block size later. Set an initial image block size (such as 5×5 pixels) as the starting point for subsequent image segmentation. Using the image segmentation coefficient (f) and the initial image block size (N0), the image block size (N) that should be used for the current frame image data is calculated by a formula. This size will be dynamically adjusted according to the complexity of the image content to optimize the processing effect. According to the calculated image block size (N), the frame image data is segmented to obtain multiple image blocks. These image blocks will serve as the basic unit for subsequent image enhancement processing. Each image block is subjected to noise suppression processing, contrast enhancement processing and sharpening processing. These processing aims to improve the visual effect of the image, make the details in the image clearer, and reduce noise interference.

[0128] The effect of the above technical solution is: through grayscale processing, noise suppression, contrast enhancement and sharpening and other processing steps, the quality of frame image data is significantly improved. This makes key information such as fire elements in the image more prominent, which is convenient for subsequent identification. The image block size is dynamically adjusted according to the complexity of the image content, making the processing process more flexible and efficient. This helps to reduce the amount of calculation while maintaining processing accuracy. After image enhancement processing, the details and features in the image are more obvious, which is conducive to the subsequent neural network model to accurately identify fire elements. Due to the improvement of image quality and the enhancement of feature extraction capabilities, the technical solution can more accurately identify fire elements in frame image data, thereby improving the accuracy of fire detection. Through automated image processing and recognition processes, intelligent monitoring and management of door and window area ranges are realized. This reduces labor costs and improves monitoring efficiency. Timely fire warnings and intelligent monitoring methods can enhance users' confidence and satisfaction with home safety. At the same time, clear and accurate image display also enhances the user's visual experience.

[0129] In one embodiment of the present invention, judging whether to perform an explosion-proof alarm according to the frame image data of the fire element includes:

[0130] S2051, extracting frame image data containing fire light elements as target frame image data;

[0131] S2052, sorting the target frame image data in order from front to back of the frame time, and obtaining an image data set corresponding to the target frame image data;

[0132] S2053, extracting the number of positions corresponding to the fire light elements contained in each target frame image data and the number of pixels contained in the fire light elements corresponding to each position;

[0133] S2054, obtaining a fire light degree evaluation coefficient corresponding to each target frame image data according to the number of positions corresponding to the fire light elements contained in each target frame image data and the number of pixels contained in the fire light elements corresponding to each position, wherein the fire light degree evaluation coefficient is obtained by the following formula:

[0134]

[0135] Where E represents the fire degree evaluation coefficient; m represents the number of target frame image data; M i Indicates the number of positions corresponding to the fire element corresponding to the i-th target frame image data; M i-1 Indicates the number of positions corresponding to the fire element corresponding to the i-1th target frame image data; B i represents the number of pixels contained in the position fire element corresponding to the jth fire element in the i-th target frame image data; B i represents the number of pixels contained in the position fire element corresponding to the j-1th fire element in the i-th target frame image data; B m Indicates the number of all pixels corresponding to the fire element of the most recent target frame image data corresponding to the frame time in the target frame image data; B c Indicates the number of all pixel points corresponding to the fire element of the first target frame image data corresponding to the frame time in the target frame image data;

[0136] S2055, sequentially comparing the fire light level evaluation coefficient corresponding to each target frame image data with a preset fire light level evaluation coefficient threshold;

[0137] S2056, when the fire light level evaluation coefficient corresponding to the target frame image data exceeds a preset fire light level evaluation coefficient threshold, an explosion-proof alarm is issued;

[0138] S2057. When the fire light level evaluation coefficients corresponding to the target frame image data that have been compared do not exceed the preset fire light level evaluation coefficient threshold, the fire light level evaluation coefficient corresponding to each target frame image data is extracted, and the fire light level evaluation coefficient corresponding to each target frame image data is used to obtain a comprehensive evaluation coefficient, wherein the comprehensive evaluation coefficient is obtained by the following formula:

[0139]

[0140] Among them, P represents the comprehensive evaluation coefficient; E mE represents the fire degree evaluation coefficient of the most recent target frame image data corresponding to the frame time in the target frame image data; y represents the preset fire light evaluation coefficient threshold; k represents the number of target frame image data that have been compared; ΔT si represents the time length between the frame time corresponding to the i-th target frame image data and the i+1-th frame time; E i represents the fire degree evaluation coefficient corresponding to the i-th target frame image data; E mc Indicates the fire light evaluation coefficient corresponding to the previous target frame image data of the most recent target frame image data; E c Indicates the fire degree evaluation coefficient corresponding to the first target frame image data;

[0141] S2058. When the comprehensive evaluation coefficient exceeds a preset comprehensive evaluation coefficient threshold, an explosion-proof alarm is issued.

[0142] The working principle of the above technical solution is as follows: extract frames containing fire elements from all frame image data as target frame image data. Sort the extracted target frame image data according to the frame time (i.e., time order) to construct an ordered image data set. Analyze the fire elements in each target frame image data, extract the number of its positions and the number of pixels corresponding to each position. This information reflects the distribution and intensity of the fire in the image. Calculate the fire degree evaluation coefficient (E) corresponding to each target frame image data by a specific formula. This coefficient comprehensively considers the number of fire positions, the number of pixels, and the changes between adjacent frames, and is used to quantify the severity of the fire. Compare the fire degree evaluation coefficient of each target frame image data with a preset threshold. If the fire degree evaluation coefficient of a frame exceeds the threshold, an explosion-proof alarm is immediately issued. If the fire degree evaluation coefficients of all compared target frame image data do not exceed the threshold, the comprehensive evaluation coefficient (P) is further calculated. This coefficient takes into account the changes in fire between multiple frames and the time factor, and is used to evaluate the comprehensive severity of the fire in the entire time period. If the comprehensive evaluation coefficient exceeds the preset threshold, an explosion-proof alarm is issued.

[0143] The effect of the above technical solution is: by comprehensively considering the number of fire positions, the number of pixels and the time changes, the technical solution can more accurately assess the severity of the fire, thereby reducing the possibility of false alarms and missed alarms. After the fire element is detected, a preliminary assessment can be made immediately, and the alarm can be quickly triggered when necessary. This helps to respond to fire incidents in a timely manner and reduce losses. By presetting different thresholds, the alarm sensitivity can be adjusted according to the needs of actual application scenarios, making the system more flexible and adaptable. Accurate fire warnings and timely alarm mechanisms can enhance users' confidence and satisfaction with home safety. At the same time, intelligent monitoring methods also provide a more convenient and efficient management method. By comprehensively evaluating the fire conditions of multiple frames of image data, the uncertainty that may be caused by a single frame of image data is reduced, and the stability and reliability of the system are improved.

[0144] The embodiment of the present invention provides a door and window explosion-proof monitoring and early warning system, such as Figure 2 As shown, the door and window explosion-proof monitoring and early warning system includes:

[0145] The first explosion-proof alarm module is used to monitor the gas concentration information and smoke concentration information corresponding to the environment where the doors and windows are located in real time through the gas sensor group and the smoke sensor, and determine whether to perform an explosion-proof alarm according to the gas concentration information and smoke concentration information corresponding to the environment where the doors and windows are located;

[0146] The second explosion-proof alarm module is used to collect image data of the door and window area in real time through an image acquisition device, and determine whether to issue an explosion-proof alarm based on the image information.

[0147] Furthermore, the first explosion-proof alarm module includes:

[0148] A gas concentration information acquisition module, used to monitor the gas concentration information corresponding to the environment of the doors and windows in real time through the gas sensor group; wherein the gas sensor group includes hydrocarbon gas sensors such as methane sensors, carbon monoxide sensors, ethane sensors and propane sensors;

[0149] A first explosion-proof alarm execution module is used to issue an explosion-proof alarm when the gas concentration information corresponding to the environment in which the doors and windows are located exceeds a preset gas concentration threshold;

[0150] A smoke concentration information acquisition module is used to monitor the smoke concentration information corresponding to the environment where the doors and windows are located in real time through the smoke concentration information;

[0151] The second explosion-proof alarm execution module is used to issue an explosion-proof alarm when the smoke concentration information corresponding to the environment in which the doors and windows are located exceeds a preset smoke concentration threshold.

[0152] The working principle of the above technical solution is: through the gas sensor group and smoke sensor deployed near the doors and windows, the gas concentration information and smoke concentration information in the environment are monitored in real time. These sensors can accurately sense the concentration changes of harmful gases (such as combustible gases, toxic gases, etc.) and smoke particles in the air. The gas concentration and smoke concentration monitored by the system in real time are compared with the preset safety threshold. If any monitoring value exceeds the threshold, the system determines that there is a safety hazard in the current environment, which may cause an explosion or fire, and then trigger an explosion-proof alarm.

[0153] Image data of the door and window area is collected in real time using an image acquisition device (such as a camera). These image data contain visual information of the doors and windows and their surroundings, and can reflect the actual situation on site. The system intelligently analyzes the collected images and identifies possible abnormal conditions, such as flames. If the image analysis results show a potential risk of explosion or fire, the system will also trigger an explosion-proof alarm.

[0154] The effect of the above technical solution is: by real-time monitoring of the gas concentration, smoke concentration and image information in the environment, the method can timely discover and warn of potential explosion or fire risks, thereby effectively avoiding the occurrence of safety accidents and ensuring the safety of personnel and property. Combined with a variety of monitoring methods, the method can monitor the environment of doors and windows from multiple dimensions. This multi-source information fusion method can significantly improve the accuracy and reliability of early warning. Once the system detects an abnormal situation and triggers an alarm, relevant departments and personnel can respond quickly and take measures to prevent the situation from further deteriorating. This rapid response mechanism is of great significance for reducing losses. The method integrates advanced technologies such as the Internet of Things and image recognition to realize intelligent management of explosion-proof monitoring of doors and windows. This not only improves monitoring efficiency, but also reduces labor costs. By providing timely and accurate early warning information, the method can enhance users' confidence and satisfaction with home safety. At the same time, intelligent monitoring methods also provide users with a more convenient and efficient management method.

[0155] In one embodiment of the present invention, the second explosion-proof alarm module comprises:

[0156] An image acquisition control module is used for the image acquisition device to acquire video image data of the door and window area in real time; wherein the image acquisition data is acquired using a high-definition camera;

[0157] A frame processing module, used for performing frame processing on the video image data to obtain frame image data corresponding to the video image data;

[0158] An image enhancement processing module is used to perform image enhancement processing on the frame image data to obtain frame image data after image enhancement processing;

[0159] A fire element judgment module is used to input the frame image data after the image enhancement processing into a neural network model for image recognition, and judge whether there is a fire element in the frame image data through the neural network model;

[0160] The explosion-proof alarm execution module is used to determine whether to issue an explosion-proof alarm based on the frame image data of the fire element.

[0161] Specifically, the neural network model structure for image recognition is as follows:

[0162] Input layer: The input layer is responsible for receiving the frame image data after image enhancement. Since the frame image data is two-dimensional (height x width), it is usually necessary to convert it into a one-dimensional vector (for example, by flattening the pixel values) or use a specific data format (such as multiple one-dimensional vectors after RGB channels are separated) as the input of the neural network.

[0163] The size of the input layer (i.e., the number of neurons) depends on the dimension and format of the frame image data. For example, if the frame image size is 640x480 pixels and is an RGB three-channel image, the input layer may need to receive input from 6404803=921600 neurons (if flattened directly), or three input vectors of size 640*480=307200 (if separated by channels).

[0164] Hidden layer: Hidden layer is the key part of neural network for feature extraction and learning. The number of hidden layers and the number of neurons in each layer can be determined according to the complexity of the specific task and the required learning ability.

[0165] In image recognition tasks, common hidden layer types include convolutional layers, pooling layers, and fully connected layers. The convolutional layer extracts local features in the image through convolution operations; the pooling layer reduces the dimension of the feature map through pooling operations, reduces the amount of calculation and extracts the main features; the fully connected layer maps the learned feature representation to the sample label space, playing the role of a classifier.

[0166] The specific configuration of the hidden layer (such as the number of layers, the number of neurons in each layer, activation function, etc.) needs to be determined through experiments and parameter adjustment to achieve the best recognition effect.

[0167] Output layer: The output layer is responsible for outputting the prediction results of the neural network. In image recognition tasks, especially in the task of determining whether there is a fire element in the frame image data, the output layer usually contains one or more neurons.

[0168] If the task is a binary classification problem (i.e., there is a fire element or there is no fire element), the output layer contains one neuron and uses the sigmoid activation function to output a probability value between 0 and 1, indicating the possibility of the presence of a fire element.

[0169] The working principle of the above technical solution is: the system uses a gas sensor group (including hydrocarbon gas sensors such as methane sensors, carbon monoxide sensors, ethane sensors and propane sensors) to monitor the gas concentration information in the environment where the doors and windows are located in real time. These sensors can accurately sense the concentration changes of specific gases in the air and provide data support for subsequent judgments.

[0170] The system compares the real-time monitored gas concentration information with the preset gas concentration thresholds. These thresholds are set according to relevant safety standards and experience to determine whether the gas concentration in the current environment is within a safe range. If any gas concentration exceeds its corresponding threshold, the system determines that there is a potential explosion risk and immediately triggers an explosion-proof alarm. While monitoring the gas concentration, the system also monitors the smoke concentration information in the environment where the doors and windows are located in real time through smoke sensors. Smoke sensors can sense smoke particles in the air and convert them into electrical signals for processing.

[0171] The system compares the real-time monitored smoke concentration information with the preset smoke concentration threshold. Similar to the gas concentration threshold, the smoke concentration threshold is also set based on safety standards and experience. If the smoke concentration exceeds its corresponding threshold, the system determines that there is a potential fire risk and also triggers the explosion-proof alarm.

[0172] The effect of the above technical solution is: by real-time monitoring of the gas concentration and smoke concentration in the environment, the method can promptly detect and warn of potential explosion or fire risks, thereby effectively avoiding the occurrence of safety accidents and ensuring the safety of personnel and property.

[0173] Using multiple sensors to monitor different parameters in the environment and combining them with preset thresholds for judgment can significantly improve the accuracy and reliability of early warning. Once the system detects an abnormal situation and triggers an alarm, relevant departments and personnel can respond quickly and take measures to prevent the situation from further deteriorating. This rapid response mechanism is of great significance for reducing losses. This method integrates the Internet of Things technology to realize intelligent monitoring and management of the environment in which doors and windows are located. This not only improves monitoring efficiency but also reduces labor costs. By adjusting the preset thresholds, the system can be flexibly configured according to different application scenarios and security requirements to adapt to different monitoring needs. By providing timely and accurate early warning information, this method can enhance users' confidence and satisfaction with home safety. At the same time, intelligent monitoring methods also provide users with a more convenient and efficient management method.

[0174] In one embodiment of the present invention, the image enhancement processing module includes:

[0175] A grayscale processing module, used to perform grayscale processing on the frame image data to obtain a grayscale value corresponding to each pixel point contained in the frame image data;

[0176] The image segmentation coefficient acquisition module is used to obtain the image segmentation coefficient according to the gray value corresponding to each pixel point contained in the frame image data, wherein the image segmentation coefficient is obtained by the following formula:

[0177]

[0178] Where, f represents the image segmentation coefficient; n represents the number of pixels contained in the frame image data; H i Indicates the gray value corresponding to the i-th pixel; H i-1 Indicates the gray value corresponding to the i-1th pixel; H max Indicates the maximum grayscale value of the pixel points contained in the frame image data; H z Indicates the central grayscale value of the pixel contained in the frame image data;

[0179] An initial image block size acquisition module, used to extract the initial image block size corresponding to the frame image data; wherein the initial image block size is 5×5;

[0180] The image block size acquisition module is used to obtain the image block size corresponding to the current frame image data by using the image segmentation coefficient combined with the initial image block size; wherein the image block size corresponding to the current frame image data is obtained by the following formula:

[0181]

[0182] Wherein, N represents the number of pixels in the horizontal and vertical directions corresponding to the image block size corresponding to the current frame image data, and N is rounded up, and the minimum value of N is 3; f represents the image segmentation coefficient; N 0 Indicates the number of pixels in the horizontal and vertical directions corresponding to the size of the initial image block;

[0183] An image segmentation module, used to segment the frame image data according to the image block size corresponding to the current frame image data, and obtain a plurality of image blocks corresponding to the current frame image data;

[0184] The block image enhancement module is used to traverse each image block corresponding to the current frame image data, and perform image enhancement processing on each image block corresponding to the current frame image data to obtain the frame image data after image enhancement processing; wherein the image enhancement processing includes noise suppression processing, contrast enhancement processing and sharpening processing.

[0185] The working principle of the above technical solution is: use a high-definition camera as an image acquisition device to capture video image data of the door and window area in real time. The high-definition camera can provide clear and delicate images, which is conducive to the accuracy and efficiency of subsequent image processing. Split the continuous video image data into separate frame image data. Frame is the basic unit of video, and each frame represents a static image. Through frame processing, the video data can be converted into a series of static image data for subsequent processing. Image enhancement processing is performed on the frame image data to improve the visual effect of the image and the accuracy of feature extraction. Image enhancement processing may include operations such as denoising, sharpening, and contrast adjustment, aiming to make the fire element in the image more prominent and easy to identify. The frame image data after image enhancement processing is input into the neural network model for image recognition. The neural network model is trained to identify specific elements in the image, such as fire. The model analyzes the input frame image and determines whether there is a fire element in it. According to the judgment result of the neural network model, if there is a fire element in the frame image data, the system believes that there is a potential fire risk in the current environment and triggers an explosion-proof alarm. The alarm signal can be carried out by sound, light signal or sending alarm information to designated equipment.

[0186] The effect of the above technical solution is: through high-definition cameras and image processing technology, image data of the door and window area can be captured in real time and clearly presented. Combined with the powerful recognition ability of the neural network model, it can accurately determine whether there are fire elements in the image, thereby improving the accuracy of fire detection. Once the fire element is detected and the alarm is triggered, the relevant departments and personnel can respond quickly and take measures to prevent the further spread of the fire. This rapid response mechanism is of great significance for reducing losses. Using the neural network model for image recognition, intelligent monitoring and management of the door and window area is realized. This not only improves the monitoring efficiency, but also reduces the labor cost. Through real-time monitoring and early warning, the method can timely detect and deal with potential fire risks, thereby enhancing the safety of people and property. Intelligent monitoring means provide users with a more convenient and efficient management method. At the same time, timely early warning information can also enhance users' confidence and satisfaction with home safety.

[0187] In one embodiment of the present invention, the explosion-proof alarm execution module includes:

[0188] A target frame image data determination module is used to extract frame image data with fire light elements as target frame image data;

[0189] An image sorting module is used to sort the target frame image data in order from front to back in frame time, and obtain an image data set corresponding to the target frame image data;

[0190] An information extraction module is used to extract the number of positions corresponding to the fire light elements contained in each target frame image data and the number of pixel points contained in the fire light elements corresponding to each position;

[0191] The fire light degree evaluation coefficient acquisition module is used to obtain the fire light degree evaluation coefficient corresponding to each target frame image data according to the number of positions corresponding to the fire light elements contained in each target frame image data and the number of pixels contained in the fire light elements corresponding to each position, wherein the fire light degree evaluation coefficient is obtained by the following formula:

[0192]

[0193] Where E represents the fire degree evaluation coefficient; m represents the number of target frame image data; M i Indicates the number of positions corresponding to the fire element corresponding to the i-th target frame image data; M i-1 Indicates the number of positions corresponding to the fire element corresponding to the i-1th target frame image data; B i represents the number of pixels contained in the position fire element corresponding to the jth fire element in the i-th target frame image data; B i represents the number of pixels contained in the position fire element corresponding to the j-1th fire element in the i-th target frame image data; B m Indicates the number of all pixels corresponding to the fire element of the most recent target frame image data corresponding to the frame time in the target frame image data; B c Indicates the number of all pixel points corresponding to the fire element of the first target frame image data corresponding to the frame time in the target frame image data;

[0194] A comparison module, used for comparing the fire light level evaluation coefficient corresponding to each target frame image data with a preset fire light level evaluation coefficient threshold in turn;

[0195] A first alarm execution module, configured to issue an explosion-proof alarm when the fire light level evaluation coefficient corresponding to the target frame image data exceeds a preset fire light level evaluation coefficient threshold;

[0196] The comprehensive evaluation coefficient acquisition module is used to extract the fire light degree evaluation coefficient corresponding to each target frame image data when the fire light degree evaluation coefficients corresponding to the target frame image data that have completed the comparison do not exceed the preset fire light degree evaluation coefficient threshold, and use the fire light degree evaluation coefficient corresponding to each target frame image data to obtain the comprehensive evaluation coefficient, wherein the comprehensive evaluation coefficient is obtained by the following formula:

[0197]

[0198] Among them, P represents the comprehensive evaluation coefficient; E m E represents the fire degree evaluation coefficient of the most recent target frame image data corresponding to the frame time in the target frame image data; y represents the preset fire light evaluation coefficient threshold; k represents the number of target frame image data that have been compared; ΔT si represents the time length between the frame time corresponding to the i-th target frame image data and the i+1-th frame time; E i represents the fire degree evaluation coefficient corresponding to the i-th target frame image data; E mc Indicates the fire light evaluation coefficient corresponding to the previous target frame image data of the most recent target frame image data; E c Indicates the fire degree evaluation coefficient corresponding to the first target frame image data;

[0199] The second alarm execution module is used to issue an explosion-proof alarm when the comprehensive evaluation coefficient exceeds a preset comprehensive evaluation coefficient threshold.

[0200] The working principle of the above technical solution is: convert the frame image data from color to grayscale image. Grayscale processing simplifies the image information, and each pixel is represented by a grayscale value, which is convenient for subsequent processing. According to the grayscale value of each pixel in the grayscale image, the image segmentation coefficient (f) is calculated by a specific formula. This coefficient takes into account the change of the grayscale value of the pixel, the maximum grayscale value (Hmax) and the central grayscale value (Hz), which is used to adjust the image block size later. Set an initial image block size (such as 5×5 pixels) as the starting point for subsequent image segmentation. Using the image segmentation coefficient (f) and the initial image block size (N0), the image block size (N) that should be used for the current frame image data is calculated by a formula. This size will be dynamically adjusted according to the complexity of the image content to optimize the processing effect. According to the calculated image block size (N), the frame image data is segmented to obtain multiple image blocks. These image blocks will serve as the basic unit for subsequent image enhancement processing. Each image block is subjected to noise suppression processing, contrast enhancement processing and sharpening processing. These processing aims to improve the visual effect of the image, make the details in the image clearer, and reduce noise interference.

[0201] The effect of the above technical solution is: through grayscale processing, noise suppression, contrast enhancement and sharpening and other processing steps, the quality of frame image data is significantly improved. This makes key information such as fire elements in the image more prominent, which is convenient for subsequent identification. The image block size is dynamically adjusted according to the complexity of the image content, making the processing process more flexible and efficient. This helps to reduce the amount of calculation while maintaining processing accuracy. After image enhancement processing, the details and features in the image are more obvious, which is conducive to the subsequent neural network model to accurately identify fire elements. Due to the improvement of image quality and the enhancement of feature extraction capabilities, the technical solution can more accurately identify fire elements in frame image data, thereby improving the accuracy of fire detection. Through automated image processing and recognition processes, intelligent monitoring and management of door and window area ranges are realized. This reduces labor costs and improves monitoring efficiency. Timely fire warnings and intelligent monitoring methods can enhance users' confidence and satisfaction with home safety. At the same time, clear and accurate image display also enhances the user's visual experience.

[0202] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. Door and window explosion-proof monitoring and early warning methods, It is characterized in that The door and window explosion-proof monitoring and early warning method comprises: The gas concentration information and smoke concentration information corresponding to the environment in which the doors and windows are located are monitored in real time by the gas sensor group and the smoke sensor, and it is determined whether to issue an explosion-proof alarm according to the gas concentration information and smoke concentration information corresponding to the environment in which the doors and windows are located; The image data of the door and window area is collected in real time by the image collection device, and the image information is used to determine whether to issue an explosion-proof alarm.

2. According to the door and window explosion-proof monitoring and early warning method of claim 1, It is characterized in that The gas concentration information and smoke concentration information corresponding to the environment of the doors and windows are monitored in real time by the gas sensor group and the smoke sensor, and whether to perform an explosion-proof alarm is determined according to the gas concentration information and smoke concentration information corresponding to the environment of the doors and windows, including: The gas sensor group is used to monitor the gas concentration information corresponding to the environment in which the doors and windows are located in real time; wherein the gas sensor group includes a methane sensor, a carbon monoxide sensor, an ethane sensor and a propane sensor; When the gas concentration information corresponding to the environment in which the doors and windows are located exceeds the preset gas concentration threshold, an explosion-proof alarm is issued; The smoke concentration information is used to monitor the smoke concentration information corresponding to the environment in which the doors and windows are located in real time; When the smoke concentration information corresponding to the environment in which the doors and windows are located exceeds a preset smoke concentration threshold, an explosion-proof alarm is issued.

3. According to the door and window explosion-proof monitoring and early warning method of claim 1, It is characterized in that The image data of the door and window area is collected in real time by an image acquisition device, and the image information is used to determine whether to perform an explosion-proof alarm, including: The image acquisition device acquires video image data of the door and window area in real time; wherein the image acquisition data is acquired using a high-definition camera; Performing frame processing on the video image data to obtain frame image data corresponding to the video image data; Performing image enhancement processing on the frame image data to obtain frame image data after image enhancement processing; Inputting the frame image data after the image enhancement processing into a neural network model for image recognition, and judging whether there is a fire element in the frame image data by the neural network model; Determine whether to issue an explosion-proof alarm based on the frame image data of the fire element.

4. According to the door and window explosion-proof monitoring and early warning method of claim 3, It is characterized in that Performing image enhancement processing on the frame image data to obtain frame image data after image enhancement processing includes: Performing grayscale processing on the frame image data to obtain a grayscale value corresponding to each pixel point contained in the frame image data; The image segmentation coefficient is obtained according to the grayscale value corresponding to each pixel point contained in the frame image data, wherein the image segmentation coefficient is obtained by the following formula: Where, f represents the image segmentation coefficient; n represents the number of pixels contained in the frame image data; H i Indicates the gray value corresponding to the i-th pixel; H i-1 Indicates the gray value corresponding to the i-1th pixel; H max Indicates the maximum grayscale value of the pixel points contained in the frame image data; H z Indicates the central grayscale value of the pixel contained in the frame image data; Extracting an initial image block size corresponding to the frame image data; wherein the initial image block size is 5×5; The image block size corresponding to the current frame image data is obtained by using the image segmentation coefficient combined with the initial image block size; wherein the image block size corresponding to the current frame image data is obtained by the following formula: Wherein, N represents the number of pixels in the horizontal and vertical directions corresponding to the image block size corresponding to the current frame image data, and N is rounded up, and the minimum value of N is 3; f represents the image segmentation coefficient; N 0 Indicates the number of pixels in the horizontal and vertical directions corresponding to the size of the initial image block; Segmenting the frame image data according to the image block size corresponding to the current frame image data to obtain a plurality of image blocks corresponding to the current frame image data; Each image block corresponding to the current frame image data is traversed, and each image block corresponding to the current frame image data is subjected to image enhancement processing to obtain the frame image data after image enhancement processing; wherein the image enhancement processing includes noise suppression processing, contrast enhancement processing and sharpening processing.

5. According to the door and window explosion-proof monitoring and early warning method of claim 3, It is characterized in that Determine whether to issue an explosion-proof alarm based on the frame image data of the fire element, including: Extracting the frame image data with the fire element as the target frame image data; Sorting the target frame image data in order from front to back of the frame time to obtain an image data set corresponding to the target frame image data; Extract the number of positions corresponding to the fire light elements contained in each target frame image data and the number of pixel points contained in the fire light elements corresponding to each position; The fire light level evaluation coefficient corresponding to each target frame image data is obtained according to the number of positions corresponding to the fire light elements contained in each target frame image data and the number of pixels contained in the fire light elements corresponding to each position, wherein the fire light level evaluation coefficient is obtained by the following formula: Where E represents the fire degree evaluation coefficient; m represents the number of target frame image data; M i Indicates the number of positions corresponding to the fire element corresponding to the i-th target frame image data; M i-1 Indicates the number of positions corresponding to the fire element corresponding to the i-1th target frame image data; B i represents the number of pixels contained in the position fire element corresponding to the jth fire element in the i-th target frame image data; B i represents the number of pixels contained in the position fire element corresponding to the j-1th fire element in the i-th target frame image data; B m Indicates the number of all pixels corresponding to the fire element of the most recent target frame image data corresponding to the frame time in the target frame image data; B c Indicates the number of all pixel points corresponding to the fire element of the first target frame image data corresponding to the frame time in the target frame image data; Comparing the fire light level evaluation coefficient corresponding to each target frame image data with a preset fire light level evaluation coefficient threshold in turn; When the fire light level evaluation coefficient corresponding to the target frame image data exceeds a preset fire light level evaluation coefficient threshold, an explosion-proof alarm is issued; When the fire light level evaluation coefficients corresponding to the target frame image data that have completed the comparison do not exceed the preset fire light level evaluation coefficient threshold, the fire light level evaluation coefficient corresponding to each target frame image data is extracted, and the fire light level evaluation coefficient corresponding to each target frame image data is used to obtain a comprehensive evaluation coefficient, wherein the comprehensive evaluation coefficient is obtained by the following formula: Among them, P represents the comprehensive evaluation coefficient; E m E represents the fire degree evaluation coefficient of the most recent target frame image data corresponding to the frame time in the target frame image data; y represents the preset fire light evaluation coefficient threshold; k represents the number of target frame image data that have been compared; ΔT si represents the time length between the frame time corresponding to the i-th target frame image data and the i+1-th frame time; E i represents the fire degree evaluation coefficient corresponding to the i-th target frame image data; E mc Indicates the fire light evaluation coefficient corresponding to the previous target frame image data of the most recent target frame image data; E c Indicates the fire light level evaluation coefficient corresponding to the first target frame image data; When the comprehensive evaluation coefficient exceeds a preset comprehensive evaluation coefficient threshold, an explosion-proof alarm is issued.

6. Door and window explosion-proof monitoring and early warning system, It is characterized in that The door and window explosion-proof monitoring and early warning system includes: The first explosion-proof alarm module is used to monitor the gas concentration information and smoke concentration information corresponding to the environment where the doors and windows are located in real time through the gas sensor group and the smoke sensor, and determine whether to perform an explosion-proof alarm according to the gas concentration information and smoke concentration information corresponding to the environment where the doors and windows are located; The second explosion-proof alarm module is used to collect image data of the door and window area in real time through an image acquisition device, and determine whether to issue an explosion-proof alarm based on the image information.

7. According to claim 6, the door and window explosion-proof monitoring and early warning system, It is characterized in that The first explosion-proof alarm module comprises: A gas concentration information acquisition module, used to monitor the gas concentration information corresponding to the environment of the doors and windows in real time through the gas sensor group; wherein the gas sensor group includes a methane sensor, a carbon monoxide sensor, an ethane sensor and a propane sensor; A first explosion-proof alarm execution module is used to issue an explosion-proof alarm when the gas concentration information corresponding to the environment in which the doors and windows are located exceeds a preset gas concentration threshold; A smoke concentration information acquisition module is used to monitor the smoke concentration information corresponding to the environment where the doors and windows are located in real time through the smoke concentration information; The second explosion-proof alarm execution module is used to issue an explosion-proof alarm when the smoke concentration information corresponding to the environment in which the doors and windows are located exceeds a preset smoke concentration threshold.

8. According to claim 6, the door and window explosion-proof monitoring and early warning system, It is characterized in that The second explosion-proof alarm module comprises: An image acquisition control module is used for the image acquisition device to acquire video image data of the door and window area in real time; wherein the image acquisition data is acquired using a high-definition camera; A frame processing module, used for performing frame processing on the video image data to obtain frame image data corresponding to the video image data; An image enhancement processing module is used to perform image enhancement processing on the frame image data to obtain frame image data after image enhancement processing; A fire element judgment module is used to input the frame image data after the image enhancement processing into a neural network model for image recognition, and judge whether there is a fire element in the frame image data through the neural network model; The explosion-proof alarm execution module is used to determine whether to issue an explosion-proof alarm based on the frame image data of the fire element.

9. According to claim 8, the door and window explosion-proof monitoring and early warning system, It is characterized in that The image enhancement processing module comprises: A grayscale processing module, used to perform grayscale processing on the frame image data to obtain a grayscale value corresponding to each pixel point contained in the frame image data; The image segmentation coefficient acquisition module is used to obtain the image segmentation coefficient according to the gray value corresponding to each pixel point contained in the frame image data, wherein the image segmentation coefficient is obtained by the following formula: Where, f represents the image segmentation coefficient; n represents the number of pixels contained in the frame image data; H i Indicates the gray value corresponding to the i-th pixel; H i-1 Indicates the gray value corresponding to the i-1th pixel; H max Indicates the maximum grayscale value of the pixel points contained in the frame image data; H z Indicates the central grayscale value of the pixel contained in the frame image data; An initial image block size acquisition module, used to extract the initial image block size corresponding to the frame image data; wherein the initial image block size is 5×5; The image block size acquisition module is used to obtain the image block size corresponding to the current frame image data by using the image segmentation coefficient combined with the initial image block size; wherein the image block size corresponding to the current frame image data is obtained by the following formula: Wherein, N represents the number of pixels in the horizontal and vertical directions corresponding to the image block size corresponding to the current frame image data, and N is rounded up, and the minimum value of N is 3; f represents the image segmentation coefficient; N 0 Indicates the number of pixels in the horizontal and vertical directions corresponding to the size of the initial image block; An image segmentation module, used to segment the frame image data according to the image block size corresponding to the current frame image data, and obtain a plurality of image blocks corresponding to the current frame image data; The block image enhancement module is used to traverse each image block corresponding to the current frame image data, and perform image enhancement processing on each image block corresponding to the current frame image data to obtain the frame image data after image enhancement processing; wherein the image enhancement processing includes noise suppression processing, contrast enhancement processing and sharpening processing.

10. According to claim 8, the door and window explosion-proof monitoring and early warning system, It is characterized in that The explosion-proof alarm execution module includes: A target frame image data determination module is used to extract frame image data with fire light elements as target frame image data; An image sorting module is used to sort the target frame image data in order from front to back in frame time, and obtain an image data set corresponding to the target frame image data; An information extraction module is used to extract the number of positions corresponding to the fire light elements contained in each target frame image data and the number of pixel points contained in the fire light elements corresponding to each position; The fire light degree evaluation coefficient acquisition module is used to obtain the fire light degree evaluation coefficient corresponding to each target frame image data according to the number of positions corresponding to the fire light elements contained in each target frame image data and the number of pixels contained in the fire light elements corresponding to each position, wherein the fire light degree evaluation coefficient is obtained by the following formula: Where E represents the fire degree evaluation coefficient; m represents the number of target frame image data; M i Indicates the number of positions corresponding to the fire element corresponding to the i-th target frame image data; M i-1 Indicates the number of positions corresponding to the fire element corresponding to the i-1th target frame image data; B i represents the number of pixels contained in the position fire element corresponding to the jth fire element in the i-th target frame image data; B i represents the number of pixels contained in the position fire element corresponding to the j-1th fire element in the i-th target frame image data; B m Indicates the number of all pixels corresponding to the fire element of the most recent target frame image data corresponding to the frame time in the target frame image data; B c Indicates the number of all pixel points corresponding to the fire element of the first target frame image data corresponding to the frame time in the target frame image data; A comparison module, used for comparing the fire light level evaluation coefficient corresponding to each target frame image data with a preset fire light level evaluation coefficient threshold in turn; A first alarm execution module, configured to issue an explosion-proof alarm when the fire light level evaluation coefficient corresponding to the target frame image data exceeds a preset fire light level evaluation coefficient threshold; The comprehensive evaluation coefficient acquisition module is used to extract the fire light degree evaluation coefficient corresponding to each target frame image data when the fire light degree evaluation coefficients corresponding to the target frame image data that have completed the comparison do not exceed the preset fire light degree evaluation coefficient threshold, and use the fire light degree evaluation coefficient corresponding to each target frame image data to obtain the comprehensive evaluation coefficient, wherein the comprehensive evaluation coefficient is obtained by the following formula: Among them, P represents the comprehensive evaluation coefficient; E m E represents the fire degree evaluation coefficient of the most recent target frame image data corresponding to the frame time in the target frame image data; y represents the preset fire light evaluation coefficient threshold; k represents the number of target frame image data that have been compared; ΔT si represents the time length between the frame time corresponding to the i-th target frame image data and the i+1-th frame time; E i represents the fire degree evaluation coefficient corresponding to the i-th target frame image data; E mc Indicates the fire light evaluation coefficient corresponding to the previous target frame image data of the most recent target frame image data; E c Indicates the fire light level evaluation coefficient corresponding to the first target frame image data; The second alarm execution module is used to issue an explosion-proof alarm when the comprehensive evaluation coefficient exceeds a preset comprehensive evaluation coefficient threshold.