Gas alarm state safety monitoring method and system based on Internet of Things, and medium

Abstract

The invention discloses a gas alarm state safety monitoring method and system based on the Internet of Things and a medium. Relates to the technical field of Internet of Things monitoring. Sending a simulation leakage signal to the gas alarm in a cycle time, and carrying out feature recognition based on the preprocessed audio-visual data to obtain recognition state data of the gas alarm; state judgment is carried out based on the recognition state data and the transmission state data of the gas alarm, and a gas alarm state judgment result is output; according to the scheme, the monitoring period is set according to the gas alarm and the environment condition, the state of the gas alarm is monitored in time, and the gas alarm is reminded of abnormity in time; the original function of the gas alarm is not changed, the state of the gas alarm is subjected to audio-visual identification by expanding and collecting audio-visual data, abnormal state judgment and centralized management are carried out by combining the transmission state data of the gas alarm, the state abnormity of the gas alarm is prompted in time, and the reliability of the combustible gas leakage monitoring process is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things monitoring, and in particular to a method, system and medium for safely monitoring the state of a gas alarm based on the Internet of Things. Background Art

[0002] Combustible gas leakage may lead to serious consequences such as fire and explosion. Gas alarms can sound an alarm at the first sign of gas leakage, reminding users to take timely measures to deal with it and effectively avoid accidents. With the development of science and technology, existing gas alarms not only have the characteristics of high precision and high sensitivity, but also can realize functions such as remote monitoring and intelligent alarm. The application of these technologies makes gas alarms play a more important role in preventing gas leaks. Therefore, for the scenario of using gas alarms for gas leak monitoring, it is very important to timely grasp whether the working status of the gas alarm is abnormal or not, so as to ensure the reliability of gas leak monitoring and avoid safety accidents.

[0003] Gas alarms mostly work in open environments and are easily interfered by external objective factors. In the process of monitoring combustible gas, there is a possibility of missed detection and false detection. Due to the lack of effective real-time monitoring methods in existing technologies, gas alarms are mainly inspected and maintained manually on a regular basis. However, manual inspection is not only costly, but also has limitations in periodicity, subjectivity and efficiency, and may not be able to detect and solve abnormal problems in a timely manner. Summary of the invention

[0004] The technical problem to be solved by the present invention is: in view of the lack of effective real-time monitoring means in the prior art, manual maintenance is not only costly, but also has limitations in periodicity, subjectivity and efficiency, and may not be able to promptly discover and solve the problem of abnormal gas alarms. The purpose of the present invention is to provide a gas alarm status safety monitoring method, system and medium based on the Internet of Things. On the one hand, a monitoring cycle is set according to the gas alarm and environmental conditions, and the status of the gas alarm is monitored in time to promptly remind the gas alarm of abnormalities; on the other hand, this scheme does not change the original function of the gas alarm, but performs audio-visual recognition of the gas alarm status by expanding the collection of audio-visual data, and judges the abnormal status in combination with the gas alarm's own transmission status data, centrally manages and promptly reminds the gas alarm of abnormal status, and increases the reliability of the combustible gas leakage monitoring process.

[0005] The present invention is achieved through the following technical solutions: With T as the period, send a simulated leakage signal to the gas alarm, collect the audiovisual data and transmission status data of the gas alarm; Preprocessing the audio-visual data; Perform feature recognition on the pre-processed audio-visual data to obtain recognition status data of the gas alarm; The status judgment is performed based on the identification status data and transmission status data of the gas alarm, and the status judgment result of the gas alarm is output.

[0006] Working principle of this solution: It provides a gas alarm status security monitoring method based on the Internet of Things. On the one hand, it sets a monitoring cycle according to the gas alarm and environmental conditions, monitors the status of the gas alarm in time, and promptly reminds the gas alarm of abnormalities; on the other hand, this solution does not change the original function of the gas alarm. It expands the collection of audio-visual data to perform audio-visual recognition of the gas alarm status, combines the gas alarm's own transmission status data to judge the abnormal status, centrally manages and promptly reminds the gas alarm of abnormal status, and increases the reliability of the combustible gas leak monitoring process.

[0007] A further optimization scheme is that the method for determining the period T includes: Obtaining attribute information of the current gas alarm; the attribute information includes: the number of maintenance times C of the gas alarm, the service life N of the gas alarm, the recommended service life Ta of the gas alarm, and the use environment index G of the gas alarm; The periodic influence factor of the current gas alarm is calculated based on the attribute information: ; Where e represents the Euler number; The monitoring period T of the current gas alarm is set according to the periodic influence factor: the factor threshold Qa is set. If the periodic influence factor Q≥factor threshold Qa, the monitoring period T is set to T1; if the periodic influence factor Q<factor threshold Qa, the monitoring period T is set to T2; wherein T1<T2.

[0008] The maintenance of traditional gas alarms mainly follows the maintenance of the units, or is repaired in accordance with national standards. Since there are usually a large number of gas alarms of different models and different usage times distributed in different regions and scenarios, if they are uniformly repaired according to the planned time, there may be a situation of untimely maintenance. This solution calculates the cycle influence factor of each gas alarm by analyzing the number of repairs of each gas alarm, the years of use of the gas alarm, the recommended service life of the gas alarm and the use environment level of the gas alarm, and sets the monitoring cycle according to the cycle influence factor. For gas alarms with hidden dangers that have many repairs, long service life and harsh use environment, a shorter status monitoring cycle plan is arranged in time, while for gas alarms with few repairs, short service life and safe use environment, a longer status monitoring plan is arranged, so as to further realize timely reminders of abnormal gas alarm status and increase the reliability of the combustible gas leakage monitoring process.

[0009] The further optimization scheme is that the method for obtaining the use environment indicators of the gas alarm includes: Obtain the current use environment parameters of the gas alarm, which include the number of surrounding leakage points n, ventilation conditions z, release source hazard level d, the average number of days D1 per year when the ambient temperature exceeds the preset temperature range, and the average number of days D2 per year when the ambient humidity exceeds the preset humidity range; The use environment index G of the gas alarm is estimated based on the use environment parameters: ; Among them, ln() represents the exponential function, and e represents the Euler number.

[0010] A further optimization scheme is to pre-process the audio-visual data; including the method: Extracting images and audio at the same moment from audio-visual data; Perform edge detection on the image to segment the gas alarm image; Grayscale masking is performed on the gas alarm image to obtain a segmented indication area image.

[0011] A further optimization scheme is that the gas alarm image is subjected to grayscale mask processing to obtain a segmented indication area image, including the following method: Configure the weight coefficients of the red channel, green channel, and blue channel of the gas alarm image, and obtain the grayscale image of the gas alarm image according to the weight coefficients; After binarizing the grayscale image, an opening operation is performed to obtain a mask image; The mask image is multiplied with the gas alarm image to obtain an indication area image.

[0012] A further optimization scheme is that the feature recognition is performed on the pre-processed audio-visual data to obtain the recognition status data of the gas alarm, including the method: Performing feature recognition on the indicated area image based on an improved nearest neighbor rule algorithm, wherein the improved nearest neighbor rule algorithm is implemented based on a k-value adaptive strategy and a feature weighting strategy; The k value adaptive strategy includes: when different new energy power generation units collect electrical quantity data to form different training samples, the algorithm will adaptively select the optimal k value and assign different weights to each feature data. The optimal k value is determined according to the test recognition accuracy P of the training sample; P=1-Y / Z; where Y represents the number of times the wrong recognition results are output in the test; and Z represents the total number of tests.

[0013] A further optimization scheme is that the feature weighting strategy includes: Calculate the Euclidean distance D based on the feature weighting strategy: ; ; ; ;

[0014] in, e i Represents the information entropy value of feature i; p ij is the jth data in feature i The proportion of N' represents the number of samples; represents the first j' individual data; x i and y i represents the coordinates of feature i; M represents the total number of features; represents the weight of feature i.

[0015] A further optimization scheme is that the state judgment is performed based on the identification state data and the transmission state data of the gas alarm, and the state judgment result of the gas alarm is output; including the method: T1, obtain the identification status data and transmission status data at the same time; T2, determine whether the current transmission status data is empty data, if so, determine that the gas alarm is in a communication failure state; otherwise, enter step T3; T3, if the expressions of the current identification status data and the current transmission status data are consistent, the gas alarm is determined to be in a normal state; if the expressions of the current identification status data and the current transmission status data are inconsistent, the gas alarm is determined to be in an abnormal state.

[0016] This solution also provides a gas alarm status safety monitoring system based on the Internet of Things, which is used to implement the above-mentioned gas alarm status safety monitoring method based on the Internet of Things. The system includes: The simulation module is used to send a simulated leakage signal to the gas alarm with T as a period; A collection module, used for collecting audio-visual data and status data of the gas alarm after the gas alarm receives a simulated leakage signal; A preprocessing module, used for preprocessing the audio-visual data; An analysis module is used to perform feature recognition on the pre-processed audio-visual data to obtain recognition status data of the gas alarm; The judgment module is used to perform status judgment based on the identification status data and status data of the gas alarm, and output the gas alarm status judgment result.

[0017] The present solution also provides a computer-readable medium on which a computer program is stored. The computer program is executed by a processor to implement the above-mentioned method for safely monitoring the status of a gas alarm based on the Internet of Things.

[0018] Compared with the prior art, the present invention has the following advantages and beneficial effects: The present invention provides a gas alarm status safety monitoring method, system and medium based on the Internet of Things. On the one hand, a monitoring cycle is set according to the gas alarm and environmental conditions, the status of the gas alarm is monitored in time, and the gas alarm abnormality is reminded in time; on the other hand, this scheme does not change the original function of the gas alarm, and the gas alarm status is audio-visually identified by expanding the collection of audio-visual data, and the abnormal status is judged in combination with the gas alarm's own transmission status data, so as to centrally manage and promptly remind the gas alarm status abnormality, thereby increasing the reliability of the combustible gas leakage monitoring process. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative work. In the drawings: Figure 1 It is a flow chart of a method for safely monitoring the status of a gas alarm based on the Internet of Things; Figure 2 This is a schematic diagram of the principle of gas alarm status safety monitoring based on the Internet of Things; Figure 3 This is a structural diagram of the gas alarm status safety monitoring system based on the Internet of Things. DETAILED DESCRIPTION

[0020] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention.

[0021] In view of the lack of effective real-time monitoring methods in the prior art, manual maintenance is not only costly, but also has limitations in periodicity, subjectivity and efficiency. It may be impossible to detect and solve abnormal problems of gas alarms in time. This solution provides the following embodiments to solve the above technical problems.

[0022] Embodiment 1: This embodiment provides a method for safely monitoring the status of a gas alarm based on the Internet of Things, such as Figure 1 and Figure 2 As shown, including: Step 1: With T as the period, send a simulated leakage signal to the gas alarm and collect the audiovisual data and transmission status data of the gas alarm; In this step, the method for determining the period T includes: S11, obtaining attribute information of the current gas alarm; the attribute information includes: the number of maintenance times C of the gas alarm, the service life N of the gas alarm, the recommended service life Ta of the gas alarm, and the use environment index G of the gas alarm; In this step, the method for obtaining the use environment index of the gas alarm includes: Obtain the current use environment parameters of the gas alarm, which include the installation location, the number of surrounding leakage points, the number of days T1 that the installation environment temperature exceeds the normal working temperature range of the gas alarm, and the number of days T2 that the installation environment humidity exceeds the normal working humidity range of the gas alarm; In this step, the method for obtaining the use environment index of the gas alarm includes: S111, obtaining the current use environment parameters of the gas alarm, the use environment parameters including the number of surrounding leakage points n, ventilation conditions z, release source hazard level d, the average number of days D1 per year when the ambient temperature exceeds the preset temperature range, and the average number of days D2 per year when the ambient humidity exceeds the preset humidity range; S112, estimating the use environment index G of the gas alarm according to the use environment parameters: ; Among them, ln() represents the exponential function, and e represents the Euler number.

[0023] S12, calculating the period influence factor of the current gas alarm based on the attribute information: ; Where e represents the Euler number; S13, set the monitoring period T of the current gas alarm according to the periodic influence factor: set the factor threshold Qa, if the periodic influence factor Q≥factor threshold Qa, then set the monitoring period T to T1; if the periodic influence factor Q<factor threshold Qa, then set the monitoring period T to T2; wherein T1<T2.

[0024] In this step, when collecting audio-visual data of the gas alarm, audio-visual collection equipment can be specially deployed according to the installation location of the gas alarm, or the existing audio-visual collection equipment in the liquid storage station or transfer station can be used; when collecting the transmission status data of the gas alarm, it can be obtained from the transmission data sent by the multi-terminal collection device of the Internet of Things to the control center.

[0025] The maintenance of traditional gas alarms mainly follows the maintenance of the unit, or is maintained in accordance with national standards; because there are usually a large number of gas alarms of different models and different usage times distributed in different regions and different scenarios, they may not be repaired in time if they are uniformly repaired according to the planned time; this solution calculates the cycle influence factor of each gas alarm by analyzing the number of repairs of each gas alarm, the years of use of the gas alarm, the recommended service life of the gas alarm, and the use environment level of the gas alarm, and sets the monitoring cycle according to the cycle influence factor. For gas alarms with hidden dangers that have many repairs, long service life, and harsh use environment, a shorter status monitoring cycle plan is arranged in time, and for gas alarms with few repairs, short service life, and safe use environment, a longer status monitoring plan is arranged, so as to further realize timely reminders of abnormal gas alarm status and increase the reliability of the combustible gas leakage monitoring process. The monitoring cycle of this solution is set outside the maintenance plan and tries not to overlap with the maintenance plan.

[0026] Step 2: Preprocess the audio-visual data; this step specifically includes the following method: S21, extracting images and audio at the same moment from the audio-visual data; S22, performing edge detection on the image to segment a gas alarm image; S23, grayscale mask processing is performed on the gas alarm image to obtain a segmented indication area image; this step specifically includes the following method: S231, configuring weight coefficients of the red channel, green channel, and blue channel of the gas alarm image, and obtaining a grayscale image of the gas alarm image according to the weight coefficients; S232, binarizing the grayscale image and then performing an opening operation to obtain a mask image; binarizing the grayscale image specifically includes a method of calculating the maximum inter-class variance of the grayscale image according to the following formula: : ; ;

[0027] Among them, p1 is the sum of the probability of occurrence of each gray value in the area from 0 to the preset threshold; m1 is the average threshold in the area from 0 to the preset threshold; p2 is the sum of the probability of occurrence of each gray value in the area from the preset threshold to 255; m2 is the average threshold in the area from the preset threshold to 255; M represents the intermediate parameter; The pixel values ​​in the grayscale image whose grayscale values ​​are greater than the maximum inter-class variance are changed to 1, and the pixel values ​​in the grayscale image whose grayscale values ​​are less than the maximum inter-class variance are changed to 0.

[0028] S233, multiplying the mask image and the gas alarm image to obtain an indication area image.

[0029] There are many components around the gas alarm, and the parts that need to be identified are mainly the gas alarms. Generally, the corresponding areas of the gas alarms have regular shapes, and there will be obvious dividing lines between the corresponding areas of the gas alarms. The image segmentation method based on edge detection can be used to realize the image segmentation of the corresponding areas of the gas alarm.

[0030] Furthermore, this solution also needs to identify the indicator light, instrument, switch and other indication areas of the gas alarm. The shape of the indication area is mainly circular, and the indication area has multiple color states. Each color state of the indication area forms the state of the gas alarm. In order to identify the state of the gas alarm, it is necessary to segment the indication area image. This solution performs grayscale masking on the gas alarm image to obtain the segmented indication area image; after grayscale masking, background segmentation is achieved, and the noise in the image is greatly reduced, making the color features extracted later more accurate, which is also conducive to improving the accuracy of the model when establishing the recognition model.

[0031] Step 3: Perform feature recognition on the pre-processed audio-visual data to obtain the recognition status data of the gas alarm; this step specifically includes the following method: Performing feature recognition on the indicated area image based on an improved nearest neighbor rule algorithm, wherein the improved nearest neighbor rule algorithm is implemented based on a k-value adaptive strategy and a feature weighting strategy; The k value adaptive strategy includes: when different new energy power generation units collect electrical quantity data to form different training samples, the algorithm will adaptively select the optimal k value and assign different weights to each feature data. The optimal k value is determined according to the test recognition accuracy P of the training sample; P=1-Y / Z; where Y represents the number of times the wrong recognition results are output in the test; and Z represents the total number of tests.

[0032] The feature weighting strategy includes: Calculate the Euclidean distance D based on the feature weighting strategy: ; ; ; ;

[0033] in, e i Represents the information entropy value of feature i; p ij is the jth data in feature i The proportion of N' represents the number of samples; represents the first j' individual data; xi and y i represents the coordinates of feature i; M represents the total number of features; represents the weight of feature i.

[0034] Since the gas alarm contains indicator lights of various colors, such as the power light is green, the alarm light is red, the fault light is yellow, the life light is yellow, etc., due to problems such as light, camera lens resolution, and camera lens focal length, there will be unclear differences in the on and off of the green and yellow indicator lights. Using a conventional indicator light recognition network will result in misidentification or failure to recognize the lights. This scheme is based on the traditional indicator light recognition method. This scheme adds a k-value adaptive strategy and a feature weighting strategy to the traditional nearest neighbor rule algorithm, adaptively selects the optimal k value, and assigns different weights to each feature data, so as to achieve key recognition of samples with unclear differences in on and off, and reduce misidentification or failure to recognize the lights.

[0035] The audiovisual data at the same moment may contain audio (such as alarm status) or may not contain audio (such as power on or power off status); therefore, this solution mainly performs feature recognition on images.

[0036] Step 4: Perform status judgment based on the identification status data and transmission status data of the gas alarm, and output the gas alarm status judgment result. This step specifically includes the following method: T1, obtain the identification status data and transmission status data at the same time; T2, determine whether the current transmission status data is empty data, if so, determine that the gas alarm is in a communication failure state; otherwise, enter step T3; T3, if the expression of the current identification status data and the current transmission status data is consistent, the gas alarm is determined to be in a normal state; if the expression of the current identification status data and the current transmission status data is inconsistent, the gas alarm is determined to be in an abnormal state. In this step, the expression of the current identification status data and the current transmission status data refers to the state of the gas alarm (power on or off, alarm on or off, fault light on, etc.) obtained according to the identification status data (power on or off, alarm on or off, fault or no fault); the state of the gas alarm (power on or off, alarm on or off, fault or no fault) obtained according to the transmission status data (power on signal or power off signal, alarm on or off signal, fault signal or no fault signal).

[0037] Embodiment 2: This embodiment provides a gas alarm status safety monitoring system based on the Internet of Things, which is used to implement the gas alarm status safety monitoring method based on the Internet of Things described in Embodiment 1, such as Figure 3As shown, the system comprises: The simulation module is used to send a simulated leakage signal to the gas alarm with T as a period; A collection module, used for collecting audio-visual data and status data of the gas alarm after the gas alarm receives a simulated leakage signal; A preprocessing module, used for preprocessing the audio-visual data; An analysis module is used to perform feature recognition on the pre-processed audio-visual data to obtain recognition status data of the gas alarm; The judgment module is used to perform status judgment based on the identification status data and status data of the gas alarm, and output the gas alarm status judgment result.

[0038] Embodiment 3: This embodiment provides a computer-readable medium on which a computer program is stored. The computer program is executed by a processor to implement the method for safely monitoring the state of a gas alarm based on the Internet of Things as described in Embodiment 1. Specifically, the following steps are performed: Step 1: With T as the period, send a simulated leakage signal to the gas alarm and collect the audiovisual data and transmission status data of the gas alarm; Step 2: preprocessing the audio-visual data; Step 3: Perform feature recognition on the pre-processed audio-visual data to obtain the recognition status data of the gas alarm; Step 4: Perform status judgment based on the identification status data and transmission status data of the gas alarm, and output the gas alarm status judgment result.

Claims

1. A gas alarm status safety monitoring method based on the Internet of Things, characterized in that: include: With time T as a cycle, send a simulated leakage signal to the gas alarm, collect audio and video data and transmission status data of the gas alarm; Preprocessing the audio-visual data; Perform feature recognition on the pre-processed audio-visual data to obtain recognition status data of the gas alarm; The status judgment is performed based on the identification status data and transmission status data of the gas alarm, and the status judgment result of the gas alarm is output.

2. The method for safely monitoring the status of a gas alarm based on the Internet of Things according to claim 1 is characterized in that: Methods for determining the period T include: Obtaining attribute information of the current gas alarm; the attribute information includes: the number of maintenance times C of the gas alarm, the service life N of the gas alarm, the recommended service life Ta of the gas alarm, and the use environment index G of the gas alarm; The periodic influence factor of the current gas alarm is calculated based on the attribute information: ; Where e represents the Euler number; The monitoring period T of the current gas alarm is set according to the periodic influence factor: the factor threshold Qa is set. If the periodic influence factor Q≥factor threshold Qa, the monitoring period T is set to T1; if the periodic influence factor Q<factor threshold Qa, the monitoring period T is set to T2; wherein T1<T2.

3. The method for safely monitoring the status of a gas alarm based on the Internet of Things according to claim 2 is characterized in that: The methods for obtaining the use environment indicators of the gas alarm include: Obtain the current use environment parameters of the gas alarm, which include the number of surrounding leakage points n, ventilation conditions z, release source hazard level d, the average number of days D1 per year when the ambient temperature exceeds the preset temperature range, and the average number of days D2 per year when the ambient humidity exceeds the preset humidity range; The use environment index G of the gas alarm is estimated based on the use environment parameters: ； Among them, ln() represents the natural logarithm function, and e represents the Euler number.

4. The method for safely monitoring the status of a gas alarm based on the Internet of Things according to claim 1 is characterized in that: Preprocessing the audio-visual data; Included methods: Extracting images and audio at the same moment from audio-visual data; Perform edge detection on the image to segment the gas alarm image; Grayscale masking is performed on the gas alarm image to obtain a segmented indication area image.

5. The method for safely monitoring the status of a gas alarm based on the Internet of Things according to claim 4 is characterized in that: The grayscale mask processing is performed on the gas alarm image to obtain a segmented indication area image, including a method: Configure the weight coefficients of the red channel, green channel, and blue channel of the gas alarm image, and obtain the grayscale image of the gas alarm image according to the weight coefficients; After binarizing the grayscale image, an opening operation is performed to obtain a mask image; The mask image is multiplied with the gas alarm image to obtain an indication area image.

6. The method for safely monitoring the status of a gas alarm based on the Internet of Things according to claim 5 is characterized in that: The method of performing feature recognition on the pre-processed audio-visual data to obtain the recognition status data of the gas alarm includes: Performing feature recognition on the indicated area image based on an improved nearest neighbor rule algorithm, wherein the improved nearest neighbor rule algorithm is implemented based on a k-value adaptive strategy and a feature weighting strategy; The k value adaptive strategy includes: when different new energy power generation units collect electrical quantity data to form different training samples, the algorithm will adaptively select the optimal k value and assign different weights to each feature data. The optimal k value is determined according to the test recognition accuracy P of the training sample; P=1-Y / Z; where Y represents the number of times an erroneous recognition result is output in the test; and Z represents the total number of tests.

7. The method for safely monitoring the status of a gas alarm based on the Internet of Things according to claim 6 is characterized in that: The feature weighting strategy includes: Calculate the Euclidean distance D based on the feature weighting strategy: ； ； ； ； in, e i Represents the information entropy value of feature i; p ij Represents the jth data in feature i The proportion of N' represents the number of samples; represents the first j' individual data; x i and y i represents the coordinates of feature i; M represents the total number of features; represents the weight of feature i.

8. The method for safely monitoring the status of a gas alarm based on the Internet of Things according to claim 1 is characterized in that: The state judgment is performed based on the identification state data and transmission state data of the gas alarm, and the state judgment result of the gas alarm is output; including method: T1, obtain the identification status data and transmission status data at the same time; T2, determine whether the current transmission status data is empty data, if so, determine that the gas alarm is in a communication failure state; Otherwise, go to step T3; T3, if the current identification state data and the current transmission state data are expressed in the same way, it is determined that the gas alarm is in a normal state; If the expressions of the current identification status data and the current transmission status data are inconsistent, it is determined that the gas alarm is in an abnormal state.

9. A gas alarm status safety monitoring system based on the Internet of Things, characterized in that: The method for safely monitoring the status of a gas alarm based on the Internet of Things is used to implement any one of claims 1 to 8, and the system comprises: The simulation module is used to send a simulated leakage signal to the gas alarm with T as a period; A collection module, used for collecting audio-visual data and status data of the gas alarm after the gas alarm receives a simulated leakage signal; A preprocessing module, used for preprocessing the audio-visual data; An analysis module is used to perform feature recognition on the pre-processed audio-visual data to obtain recognition status data of the gas alarm; The judgment module is used to perform status judgment based on the identification status data and status data of the gas alarm, and output the gas alarm status judgment result.

10. A computer readable medium having a computer program stored thereon, characterized in that: The computer program is executed by a processor to implement the gas alarm status safety monitoring method based on the Internet of Things as described in any one of claims 1 to 8.

Images