Comprehensive monitoring and early warning system and method for fire leaving in kitchen
By obtaining and analyzing kitchen data in real time, the alarm time is dynamically corrected, which solves the limitations of fixed time warning in the existing technology, and achieves more accurate fire warnings, which significantly improves kitchen safety.
Patent Information
- Application Number
- CN202411190052.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-11-15
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing fire-off warning controller sets a fixed time during early warning, and does not consider the risk coefficient in actual scenarios, resulting in insufficient accuracy of the warning and certain limitations.
By obtaining the image, temperature and humidity data of the target area of the kitchen in real time, building multiple functional relationships, calculating the rate of change and abnormal coefficients, dynamically correcting the alarm time, so as to accurately judge the phenomenon of fire leaving the person and issue an early warning.
It improves the accuracy and safety of early warnings, and can automatically take emergency measures at critical moments to effectively prevent fire accidents and significantly improve the safety level of kitchens and similar areas.
Smart Images

Figure CN118968701A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of comprehensive early warning monitoring, and in particular to a comprehensive monitoring and early warning system and method for kitchen fire and absence of personnel. Background Art
[0002] In the modern catering industry, commercial kitchens are a highly active and potentially hazardous working environment, involving frequent hot work and the use of flammable substances. Given these conditions, the incidence of commercial kitchen fires is relatively high, which may not only result in significant casualties and property losses, but also cause serious disruption to catering operations.
[0003] In view of this situation, in addition to strengthening fire prevention awareness and equipping fire-fighting tools, it is more practical and effective to take measures to prevent fires from happening through supervision and prevention and control. The fire alarm system is a fire warning and response system that combines detection sensing technology and IoT intelligent control technology. It responds by sensing the activity range and activity time of the chef in the kitchen fire area, and adopts sound and light alarm functions to warn the chef to return to the operation area in time. At the same time, when the warning danger occurs, the fire warning controller can feed back the warning signal to the fire control center. However, since the existing fire warning controller sets a fixed time for warning, it does not consider the risk factor in the actual scenario and has certain limitations. Summary of the invention
[0004] The purpose of the present invention is to provide a kitchen fire and absence comprehensive monitoring and early warning system and method to solve the following technical problems: At the same time, when a warning danger occurs, the fire and occupant warning controller can feed back the warning signal to the fire control center. However, since the existing fire and occupant warning controller is set to issue a warning at a fixed time, it does not take into account the danger factor in the actual scenario and has certain limitations.
[0005] The purpose of the present invention can be achieved through the following technical solutions: A kitchen fire and absence comprehensive monitoring and early warning method, comprising the following steps: S1, acquiring monitoring data of a target area in real time, wherein the monitoring data includes image data, temperature data and humidity data; S2, analyzing the image monitoring data to identify whether there is a phenomenon of fire and absence of people in the current target area, if there is a phenomenon of fire and absence of people, extracting the temperature data and humidity data of the target area in the past time period T, preprocessing the temperature data and humidity data, and constructing a first image with time as the horizontal coordinate and temperature data as the vertical coordinate based on the preprocessed temperature data and humidity data, and determining the functional relationship W(t) of the first target image; constructing a second image with time as the horizontal coordinate and humidity data as the vertical coordinate, and determining the functional relationship S(t) of the second target image; Preset observation time h and alarm time H, obtain temperature data and humidity data in real time, construct a third image with time as the horizontal coordinate and temperature data as the vertical coordinate, and determine the functional relationship W'(t) of the third target image; construct a fourth image with time as the horizontal coordinate and humidity data as the vertical coordinate, and determine the functional relationship S'(t) of the fourth target image; wherein h<H; S3, calculate the first change rate , the second rate of change , the third rate of change and the fourth rate of change , calculate the abnormality coefficient ε according to the change rate, and correct the alarm time H according to the abnormality coefficient.
[0006] As a further solution of the present invention: in S2, the specific process of pretreatment is: Detect outliers in the data based on the superior-inferior solution distance method, delete the detected outliers, and correct the data through the interpolation algorithm As a further solution of the present invention: In S2, the specific identification process of judging the phenomenon of fire and separation is: The image data is input into the SSD target detection algorithm for detection. If a detection frame exists, it is determined that a target person exists. If no detection frame exists, a polygon intersection and ratio calculation is performed based on the smoke concentration exceeding standard image group, the stove firing image and the real-time image. If the ratio is greater than a preset first threshold, it is determined that a fire is started and a person is away from the stove.
[0007] As a further solution of the present invention: S2 also includes calculating the polygon intersection and union ratio in real time within the alarm time based on the smoke concentration exceeding standard image group, the stove firing image and the real-time image; if the intersection and union ratio is greater than or equal to a second preset threshold, the alarm time H is corrected to the current moment.
[0008] As a further solution of the present invention: in S3, the calculation formula of the abnormal coefficient is:
[0009] As a further solution of the present invention: the specific process of correcting the alarm time H is: The abnormal coefficient ε is obtained. If ε is a positive number, the alarm time is not corrected; if ε is a negative number, the corrected alarm time H' is calculated according to the correction formula H'=H*(1+ε / C'a).
[0010] As a further solution of the present invention: if the corrected alarm time H' is less than or equal to the observation time h, an alarm is directly issued.
[0011] As a further solution of the present invention: if the corrected alarm time H' is greater than the observation time h, then calculate If t' is greater than 0, then take the current time as the starting point and repeat S2 and S3.
[0012] A kitchen fire and absence comprehensive monitoring and early warning system, characterized by comprising: A data acquisition module, which acquires monitoring data of a target area in real time, wherein the monitoring data includes image data, temperature data and humidity data; A data analysis module analyzes the image monitoring data to identify whether there is a phenomenon of fire and no people in the current target area. If there is a phenomenon of fire and no people, the temperature data and humidity data of the target area in the past time period T are extracted, and the temperature data and humidity data are preprocessed. According to the preprocessed temperature data and humidity data, a first image is constructed with time as the horizontal coordinate and temperature data as the vertical coordinate, and a functional relationship W(t) of the first target image is determined; a second image is constructed with time as the horizontal coordinate and humidity data as the vertical coordinate, and a functional relationship S(t) of the second target image is determined; Preset observation time h and alarm time H, obtain temperature data and humidity data in real time, construct a third image with time as the horizontal coordinate and temperature data as the vertical coordinate, and determine the functional relationship W'(t) of the third target image; construct a fourth image with time as the horizontal coordinate and humidity data as the vertical coordinate, and determine the functional relationship S'(t) of the fourth target image; wherein h<H; Result generation module, calculates the first rate of change , the second rate of change , the third rate of change and the fourth rate of change , calculate the abnormality coefficient ε according to the change rate, and correct the alarm time H according to the abnormality coefficient.
[0013] Beneficial effects of the present invention: The present invention acquires monitoring data of the target area in real time, pre-processes the acquired temperature data and humidity data, and constructs a functional relationship between time and temperature and humidity based on these data. By calculating the rate of change and the abnormal coefficient ε, the system can accurately determine whether there is an abnormal situation, and dynamically correct the alarm time H according to the actual risk level, thereby improving the accuracy of the early warning. Through real-time monitoring and accurate early warning, the fire risk caused by the phenomenon of leaving the fire is effectively solved, and the safety level of kitchens and similar areas is significantly improved. Taking into account the risk factor in actual scenarios, emergency measures can be automatically taken at critical moments to effectively prevent accidents. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] The present invention will be further described below in conjunction with the accompanying drawings.
[0015] Figure 1 The invention discloses a flow chart of a comprehensive monitoring and early warning method for kitchen fire and absence of personnel. DETAILED DESCRIPTION
[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0017] See also Figure 1 As shown, the present invention is a comprehensive monitoring and early warning method for kitchen fire and absence of personnel, comprising the following steps: S1, acquiring monitoring data of a target area in real time, wherein the monitoring data includes image data, temperature data and humidity data; S2, analyzing the image monitoring data to identify whether there is a phenomenon of fire and absence of people in the current target area, if there is a phenomenon of fire and absence of people, extracting the temperature data and humidity data of the target area in the past time period T, preprocessing the temperature data and humidity data, and constructing a first image with time as the horizontal coordinate and temperature data as the vertical coordinate based on the preprocessed temperature data and humidity data, and determining the functional relationship W(t) of the first target image; constructing a second image with time as the horizontal coordinate and humidity data as the vertical coordinate, and determining the functional relationship S(t) of the second target image; Preset observation time h and alarm time H, obtain temperature data and humidity data in real time, construct a third image with time as the horizontal coordinate and temperature data as the vertical coordinate, and determine the functional relationship W'(t) of the third target image; construct a fourth image with time as the horizontal coordinate and humidity data as the vertical coordinate, and determine the functional relationship S'(t) of the fourth target image; wherein h<H; S3, calculate the first change rate , the second rate of change , the third rate of change and the fourth rate of change , calculate the abnormality coefficient ε according to the change rate, and correct the alarm time H according to the abnormality coefficient.
[0018] The present invention acquires monitoring data of the target area in real time, pre-processes the acquired temperature data and humidity data, and constructs a functional relationship between time and temperature and humidity based on these data. By calculating the rate of change and the abnormal coefficient ε, the system can accurately determine whether there is an abnormal situation, and dynamically correct the alarm time H according to the actual risk level, thereby improving the accuracy of the early warning. Through real-time monitoring and accurate early warning, the fire risk caused by the phenomenon of leaving the fire is effectively solved, and the safety level of kitchens and similar areas is significantly improved. Taking into account the risk factor in actual scenarios, emergency measures can be automatically taken at critical moments to effectively prevent accidents.
[0019] It is worth noting that the prior art fire-starting and leaving-personnel technology first sets the warning time, alarm time and warning time. When the set warning time is reached, a voice, sound and light warning is automatically given to warn the indoor personnel to return to the kitchen to operate or supervise in time; when the kitchen fire starts and the person is away from the kitchen reaches the warning time, an alarm is automatically sent to the user's mobile phone through the transmission network, and at the same time, the automatic alarm linkage closes the kitchen's gas emergency shut-off valve to cut off the kitchen gas source, so as to prevent the kitchen fire, gas leakage and other safety accidents caused by too long of leaving the person from the fire or other unexpected reasons. However, since the alarm time is pre-set and does not take into account the risk factor in the actual scenario, there is a problem. It has certain limitations. For example, in some high-risk environments, a shorter alarm time may be required to ensure safety. It is understandable that when there is a fire and no one is around, the gas valve of the stove is fixed. Therefore, the temperature in the kitchen usually rises until it is balanced. It is understandable that normal activities in the kitchen, such as cooking and making soup, will cause the humidity in the kitchen to rise. When there is a fire and no one is around, the temperature and humidity change rate in the kitchen can be obtained to determine the current danger factor of the kitchen, so as to adjust the alarm time according to the danger factor. The danger factor in the actual scenario is taken into consideration, which effectively prevents the occurrence of accidents and improves regional safety.
[0020] In another preferred embodiment of the present invention, in S2, the specific process of pre-processing is: The outliers in the data are detected based on the superior-inferior solution distance method, the detected outliers are deleted, and the data are corrected through the interpolation algorithm.
[0021] In another preferred embodiment of the present invention, in S2, the specific identification process of judging the phenomenon of fire and separation is: The image data is input into the SSD target detection algorithm for detection. If a detection frame exists, it is determined that a target person exists. If no detection frame exists, a polygon intersection and ratio calculation is performed based on the smoke concentration exceeding standard image group, the stove firing image and the real-time image. If the ratio is greater than a preset first threshold, it is determined that a fire is started and a person is away from the stove.
[0022] In another preferred embodiment of the present invention, S2 also includes calculating the polygon intersection and union ratio in real time within the alarm time based on the smoke concentration exceeding standard image group, the stove firing image and the real-time image; if the intersection and union ratio is greater than or equal to the second preset threshold, the alarm time H is corrected to the current moment.
[0023] In another preferred embodiment of the present invention, in S3, the calculation formula of the abnormal coefficient is:
[0024] In another preferred embodiment of the present invention, the specific process of correcting the alarm time H is as follows: The abnormal coefficient ε is obtained. If ε is a positive number, the alarm time is not corrected; if ε is a negative number, the corrected alarm time H' is calculated according to the correction formula H'=H*(1+ε / C'a).
[0025] In another preferred embodiment of the present invention, if the corrected alarm time H' is less than or equal to the observation time h, an alarm is directly issued.
[0026] In another preferred embodiment of the present invention, if the corrected alarm time H' is greater than the observation time h, then the calculation If t' is greater than 0, then take the current time as the starting point and repeat S2 and S3.
[0027] A kitchen fire and absence comprehensive monitoring and early warning system, comprising: A data acquisition module, which acquires monitoring data of a target area in real time, wherein the monitoring data includes image data, temperature data and humidity data; A data analysis module analyzes the image monitoring data to identify whether there is a phenomenon of fire and no people in the current target area. If there is a phenomenon of fire and no people, the temperature data and humidity data of the target area in the past time period T are extracted, and the temperature data and humidity data are preprocessed. According to the preprocessed temperature data and humidity data, a first image is constructed with time as the horizontal coordinate and temperature data as the vertical coordinate, and a functional relationship W(t) of the first target image is determined; a second image is constructed with time as the horizontal coordinate and humidity data as the vertical coordinate, and a functional relationship S(t) of the second target image is determined; Preset observation time h and alarm time H, obtain temperature data and humidity data in real time, construct a third image with time as the horizontal coordinate and temperature data as the vertical coordinate, and determine the functional relationship W'(t) of the third target image; construct a fourth image with time as the horizontal coordinate and humidity data as the vertical coordinate, and determine the functional relationship S'(t) of the fourth target image; wherein h<H; Result generation module, calculates the first rate of change , the second rate of change , the third rate of change and the fourth rate of change , calculate the abnormality coefficient ε according to the change rate, and correct the alarm time H according to the abnormality coefficient.
[0028] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.
Claims
1. A kitchen fire and absence comprehensive monitoring and early warning method, characterized in that: The following steps are involved: S1, acquiring monitoring data of a target area in real time, wherein the monitoring data includes image data, temperature data and humidity data; S2, analyzing the image monitoring data to identify whether there is a phenomenon of fire and absence of people in the current target area, if there is a phenomenon of fire and absence of people, extracting the temperature data and humidity data of the target area in the past time period T, preprocessing the temperature data and humidity data, and constructing a first image with time as the horizontal coordinate and temperature data as the vertical coordinate based on the preprocessed temperature data and humidity data, and determining the functional relationship W(t) of the first target image; constructing a second image with time as the horizontal coordinate and humidity data as the vertical coordinate, and determining the functional relationship S(t) of the second target image; Preset observation time h and alarm time H, obtain temperature data and humidity data in real time, construct a third image with time as the horizontal coordinate and temperature data as the vertical coordinate, and determine the functional relationship W'(t) of the third target image; construct a fourth image with time as the horizontal coordinate and humidity data as the vertical coordinate, and determine the functional relationship S'(t) of the fourth target image; wherein h<H; S3, calculate the first change rate , the second rate of change , the third rate of change and the fourth rate of change , calculate the abnormality coefficient ε according to the change rate, and correct the alarm time H according to the abnormality coefficient.
2. A kitchen fire and absence comprehensive monitoring and early warning method according to claim 1, characterized in that: In S2, the specific process of preprocessing is: The outliers in the data are detected based on the superior-inferior solution distance method, the detected outliers are deleted, and the data are corrected through the interpolation algorithm.
3. A kitchen fire and absence comprehensive monitoring and early warning method according to claim 1, characterized in that: In S2, the specific identification process of judging the phenomenon of fire and human separation is as follows: The image data is input into the SSD target detection algorithm for detection. If a detection frame exists, it is determined that a target person exists. If no detection frame exists, a polygon intersection and ratio calculation is performed based on the smoke concentration exceeding standard image group, the stove firing image and the real-time image. If the ratio is greater than a preset first threshold, it is determined that a fire is started and a person is away from the stove.
4. A kitchen fire and absence comprehensive monitoring and early warning method according to claim 1, characterized in that: The S2 also includes calculating the polygon intersection and union ratio in real time within the alarm time based on the smoke concentration exceeding standard image group, the stove firing image and the real-time image. If the intersection and union ratio is greater than or equal to the second preset threshold, the alarm time H is corrected to the current moment.
5. A kitchen fire and absence comprehensive monitoring and early warning method according to claim 1, characterized in that: In S3, the calculation formula of the abnormal coefficient is:
6. A kitchen fire and absence comprehensive monitoring and early warning method according to claim 1, characterized in that: The specific process of correcting the alarm time H is as follows: The abnormal coefficient ε is obtained. If ε is a positive number, the alarm time is not corrected; if ε is a negative number, the corrected alarm time H' is calculated according to the correction formula H'=H*(1+ε / C'a).
7. A kitchen fire and absence comprehensive monitoring and early warning method according to claim 6, characterized in that: If the corrected alarm time H' is less than or equal to the observation time h, an alarm is directly issued.
8. A kitchen fire-related integrated monitoring and early warning method according to claim 1, characterized in that: If the corrected alarm time H' is greater than the observation time h, then calculate If t' is greater than 0, then take the current time as the starting point and repeat S2 and S3.
9. A kitchen fire and absence comprehensive monitoring and early warning system, characterized in that: include: A data acquisition module, which acquires monitoring data of a target area in real time, wherein the monitoring data includes image data, temperature data and humidity data; A data analysis module analyzes the image monitoring data to identify whether there is a phenomenon of fire and no people in the current target area. If there is a phenomenon of fire and no people, the temperature data and humidity data of the target area in the past time period T are extracted, and the temperature data and humidity data are preprocessed. According to the preprocessed temperature data and humidity data, a first image is constructed with time as the horizontal coordinate and temperature data as the vertical coordinate, and a functional relationship W(t) of the first target image is determined; a second image is constructed with time as the horizontal coordinate and humidity data as the vertical coordinate, and a functional relationship S(t) of the second target image is determined; Preset observation time h and alarm time H, obtain temperature data and humidity data in real time, construct a third image with time as the horizontal coordinate and temperature data as the vertical coordinate, and determine the functional relationship W'(t) of the third target image; construct a fourth image with time as the horizontal coordinate and humidity data as the vertical coordinate, and determine the functional relationship S'(t) of the fourth target image; wherein h<H; Result generation module, calculates the first rate of change , the second rate of change , the third rate of change and the fourth rate of change , calculate the abnormality coefficient ε according to the change rate, and correct the alarm time H according to the abnormality coefficient.
Citation Information
Cited By
Kitchen safety detection method and alarm
CN121438507A
