KR20220155710A - System and method for management of hazardous gas - Google Patents

Abstract

According to the present invention, a harmful gas management system includes: at least one gas sensor for forming a gas detection signal when a harmful gas is generated; a server for receiving a gas detection signal and forming a danger warning signal; and one or more safety equipment equipped with an alarm unit which is worn by a worker and receives a danger alarm signal and informs generation of the harmful gas using sound and/or light.

Description

Hazardous gas management system and method {SYSTEM AND METHOD FOR MANAGEMENT OF HAZARDOUS GAS}

The present invention relates to a harmful gas management system and method, by installing a gas sensor in a place where harmful gas can occur, immediately transmitting a signal to a worker's safety equipment when harmful gas is generated, and detecting harmful gas by using sound and/or light. It relates to a harmful gas management system and method that can notify.

Workers (construction workers, firefighters, shipyard workers, etc.) who are put into various work sites (construction work sites, firefighting work sites, shipyards, etc.) are exposed to various risk factors, especially harmful gases at closed work sites. In the event of a leakage accident, it acts as a fatal risk to the safety of workers.

Although a gas sensor that can simply detect the leakage of harmful gas has been installed at the work site in the past, there is no means to effectively notify workers in the event of a harmful gas leak, or an alarm sound generated by the gas sensor itself When the sound of is small, there is a problem in that workers present at a distance away from the gas sensor cannot hear the alarm sound.

Korean Patent Publication No. 10-2006-0068574 (portable electrochemical gas sensor equipped with communication module, published on June 21, 2006)

The present invention provides a harmful gas management system and method capable of notifying the occurrence of harmful gas using sound and/or light by immediately transmitting a signal to the operator's safety equipment when a harmful gas is generated by installing a gas sensor in a place where harmful gas can occur. to provide.

Harmful gas management system according to an embodiment of the present invention, one or more gas sensors for forming a gas detection signal when harmful gas is generated; a server configured to receive the gas detection signal and form a danger warning signal; and one or more safety equipment worn by a worker and equipped with an alarm unit for receiving the danger warning signal and notifying the occurrence of the noxious gas using at least one of sound and light.

In addition, the one or more safety equipment may include any one or more of a helmet, safety shoes, a safety vest, a fire protection suit, a safety belt, and an oxygen cylinder for extinguishing a fire.

In addition, the one or more safety devices may further include a heart rate sensor that forms heart rate information or breathing information of the operator.

In addition, it may further include a photographing unit for acquiring a face image of the worker working at the work site at a location spaced apart from the worker.

In addition, the server extracts a pupil object by applying the face image to a pre-learned neural network, and determines whether the worker is looking at a work object at the work site using the pupil object, and the worker A warning signal may be generated when it is determined that the user is not looking at the work object.

In addition, the server extracts the coordinates of the most protruding pupil part based on the face of the operator from the pupil object, generates a perpendicular line from the extracted coordinates, and the generated perpendicular line is the object of the work object. If it does not intersect with the work object, it may be determined that the operator is not looking at the work object, and if the generated repair line intersects the work object, it may be determined that the worker is looking at the work object.

In addition, the server obtains training face images, extracts training feature point objects from the training face images, obtains labels that are body part information corresponding to the training feature point objects, and converts the training feature point objects into the neural network. Applying to a network, training outputs corresponding to the training feature point objects may be generated, and the pupil object may be extracted by learning the neural network based on the training outputs and the labels.

Also, the gas sensor may detect the concentration of the noxious gas, and the server may differently form the danger warning signal according to the concentration of the noxious gas.

In addition, a location recognizing means capable of detecting the location of the operator is installed in the one or more safety devices, and the server determines the location where the gas sensor that forms the gas detection signal is installed and the operator identified through the location recognizing means. The danger alert signal may be transmitted by comparing the positions of the operator to select the operator within a predetermined distance.

A harmful gas management method according to another embodiment of the present invention includes forming a gas detection signal when harmful gas is generated by one or more gas sensors; generating, by a server, a danger warning signal by receiving the gas detection signal; and receiving the danger alarm signal by an alarm unit provided in one or more safety equipment worn by a worker and notifying the occurrence of the noxious gas using at least one of sound and light.

Harmful gas management system according to another embodiment of the present invention, one or more gas sensors for forming a gas detection signal when harmful gas is generated; a server configured to receive the gas detection signal and form a danger warning signal; and an alarm unit receiving the danger warning signal and notifying the occurrence of the noxious gas using at least one of sound and light, wherein the gas sensor and the alarm unit are installed in one or more safety equipment worn by a worker.

In addition, the gas sensor may further generate its own alarm signal when the noxious gas is generated.

According to embodiments of the present invention, it is possible to prevent industrial accidents that may occur at work sites due to harmful gases. Accordingly, loss of life and economic damage due to industrial accidents can be reduced.

1 is an exemplary view showing a harmful gas management system according to an embodiment of the present invention.
2 is an exemplary view showing the configuration of a server according to an embodiment of the present invention.
3 is a diagram for explaining learning of a neural network according to an embodiment of the present invention.
Figure 4 is a flow chart showing the procedure of the harmful gas management method according to an embodiment of the present invention.

Embodiments of the present invention are illustrated for the purpose of explaining the technical idea of the present invention. The scope of rights according to the present invention is not limited to the specific description of the embodiments or these embodiments presented below.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

The harmful gas management system according to the present invention includes one or more gas sensors 112 for distinguishing harmful gas from non-harmful gas and forming a gas detection signal when harmful gas is generated, and a server for receiving the gas detection signal and forming a danger alarm signal ( 114), and one or more safety equipment 122 worn by the operator 120 and equipped with an alarm unit 124 for receiving a danger alarm signal and notifying the occurrence of harmful gas using sound and / or light, An object of the present invention is to provide a harmful gas management system capable of reducing human loss and economic damage due to industrial accidents.

1 is an exemplary view showing a harmful gas management system according to an embodiment of the present invention.

As shown in FIG. 1, the harmful gas management system 100 includes a work site 110, a gas sensor 112, a server 114, a worker 120, safety equipment 122, and an alarm unit 124. And it may include a photographing unit 130.

First, the work site 110 is an area where certain work is performed, such as a construction work site such as architecture / civil engineering, a fire suppression work site, a shipyard, and a production line site in a factory, and includes both open and closed sites, The worker 120 means a person who performs a certain task at the work site 110 .

Next, the gas sensor 112 distinguishes between harmful gas and non-toxic gas and forms a gas detection signal when detecting the generation of harmful gas, and there is no particular limitation as long as it is a gas sensor capable of detecting harmful gas. In addition, harmful gases are gases that can adversely affect the human body, and include sulfur oxides (SOx), nitrogen oxides (NOx), oxides, hydrocarbons, fluorine compounds, carbon monoxide, carbon dioxide, ammonia, etc., but are limited thereto. It is not.

In addition, the gas sensor 112 may be installed at one or more work sites 110, and preferably, a plurality of gas sensors 112 may be installed at predetermined intervals within the work site 110, and noxious gas may be generated. In addition, it is preferable to configure to detect the concentration of harmful gases together. In addition, depending on the harmful gas to be detected, it is preferable to install it on the upper side of the work site 110 in the case of a harmful gas lighter than air, and to install it on the lower side of the work site 110 in the case of a harmful gas heavier than air. desirable.

In addition, the gas sensor 112 may be configured to generate its own alarm signal in addition to forming a gas detection signal when it detects the generation of harmful gas, and may use sound and/or light as its own alarm signal. .

Next, the server 114 may receive a gas detection signal from the gas sensor 112 to form a danger warning signal. According to one embodiment, the server 114, when receiving a gas detection signal from the gas sensor 112 may form a danger warning signal for recognizing it to the worker 120.

Next, the alarm unit 124 may receive a danger warning signal from the server 114 and notify the generation of noxious gas using sound and/or light. According to one embodiment, the alarm unit 124 receives a danger alarm signal from the server 114 and generates an alarm signal in the form of a sound, and/or generates a siren in the form of light to alert the operator ( 120) can be notified. For example, the alarm unit 124 may be installed in the safety equipment 122 worn by the worker 120 .

In addition, the alarm unit 124 may receive, from the server 114, a danger warning signal formed differently according to the concentration of the harmful gas detected by the gas sensor 112, and change the sound and/or light alarm method. According to one embodiment, the gas sensor 112 may detect the concentration of the harmful gas, and the server 114 may set a different danger alarm signal according to the concentration of the harmful gas, for example, according to the concentration of the harmful gas. , the danger alert signal can be formed by changing the frequency of the alarm sound or changing the color of the light. Due to this, the worker 120 can immediately know the concentration of the harmful gas only with the form of the alarm, such as the frequency of the alarm sound or the color of the light, so that the worker 120 can flexibly cope with the concentration of the harmful gas. .

Next, the safety equipment 122 is a protective equipment that can be worn by the worker 120 at the work site 110, and may include, but is not limited to, a helmet, safety shoes, a safety vest, a fire protection suit, a safety belt, an oxygen cylinder for fire suppression, and the like. .

In addition, the safety device 122 may further include a location recognition unit, a communication unit, and a heartbeat sensor, although not shown, in addition to the above-described alarm unit 124 .

At this time, the location recognition means is attached to the safety equipment 122 to form the location and movement route information of the worker 120 wearing the safety equipment 122, a GPS (Global Positioning System) receiver, Galileo ) receiver, Glonass receiver, etc., but is not limited thereto.

In addition, the communication unit is capable of receiving the danger alert signal formed by the server 114, and may include a Bluetooth module, a mobile communication module, an infrared communication module, etc., but is not limited thereto, eMBB (enhanced mobile broadband), URLLC (Ultra Reliable Low-Latency Communications), MMTC (Massive Machine Type Communications), LTE (long-term evolution), LTE-A (LTE Advance), UMTS (Universal Mobile Telecommunications System), GSM (Global System for Mobile communications) , code division multiple access (CDMA), wideband CDMA (WCDMA), wireless broadband (WiBro), wireless fidelity (WiFi), Bluetooth, near field communication (NFC), global positioning system (GPS) or global navigation (GNSS) satellite system).

Alternatively, the communication unit may perform wired communication according to a method such as universal serial bus (USB), high definition multimedia interface (HDMI), recommended standard 232 (RS-232), or plain old telephone service (POTS).

In addition, the heart rate sensor may be attached to the safety equipment 122 to form heart rate information or breathing information of the worker 120 .

On the other hand, the server 114 may receive movement path information of the worker 120 from the above-described location recognition means to determine the location and moving line of the worker 120 wearing the safety equipment 122, at this time The server 114 may determine the priority of the worker 120 to transmit the danger alert signal based on the location and moving movement of the worker 120 received from the above-described location recognition means. That is, the server 114 determines the location and movement of the worker 120 through the location recognition means installed in the safety equipment 122 of the worker 120 and the location where the gas sensor 112 that forms the gas detection signal is installed. By comparing route information, a danger alert signal may be transmitted by selecting a worker 120 within a predetermined distance (eg, 1m, 2m, etc.).

Specifically, the server 114 may preferentially transmit the danger alert signal to the worker 120 located in a location close to the location where the gas sensor 112 generating the gas detection signal is installed. According to one embodiment, when a plurality of gas sensors 112 are installed in various locations, a specific gas sensor 112 forming a gas detection signal among these gas sensors 112 is installed and an operator receives information from the location recognition unit. As the position and movement path information of 120 is located in a short distance, a danger alert signal may be transmitted to the worker 120 with priority. As a result, even when there are a plurality of workers 120, communication traffic can be solved by prioritizing the danger alert signal, and at the same time, the safety of the workers 120 directly exposed to danger can be secured with priority.

In addition, according to another embodiment, when a plurality of gas sensors 112 are installed in various locations, a specific gas sensor 112 forming a gas detection signal among these gas sensors 112 is installed and received from the location recognizing means. When the location and movement path information of one worker 120 is located within the first distance, the server 114 transmits a first danger alert signal to the worker 120 located within the first distance, and The server 114 may transmit a second danger alert signal to the worker 120 . In this case, the first distance may be shorter than the second distance, and the operator 120 located within the first distance may be closer to the location where the specific gas sensor 112 is installed. The first danger alert signal and the second danger alert signal may be respective danger alert signals set to have alarm sound frequencies or light colors different. Due to this, even when there are a plurality of workers 120, different danger alarm signals are generated according to the proximity to the specific gas sensor 112 forming the gas detection signal, thereby promptly evacuating the worker 120 according to the degree of exposure to danger. can induce

In addition, the server 114, based on the location and movement path information of the worker 120 received from the location recognition means, the worker 120 is a harmful gas dangerous area (location where a specific gas sensor forming a gas detection signal is installed) When it is determined that it is out of range, the operation of the alarm unit 124 can be stopped.

In addition, the server 114 monitors the heart rate information or breathing information of the worker 120 formed by the heart rate sensor of the safety equipment 122 to rescue the worker 120 when the worker 120 is put into a closed work site or a fire suppression work site. Accidents involving human life can be prevented in advance through prompt input of manpower.

Next, the photographing unit 130 may acquire a face image of the worker 120 working at the work site 110 at a location spaced apart from the worker 120 . According to an embodiment, the photographing unit 130 may transmit the acquired face image to the server 114 .

Meanwhile, in the case of the above-described embodiment, it is assumed that the gas sensor 112 is installed at the work site 110, but other embodiments in which the gas sensor 112 is directly installed in the safety equipment 122 of the operator 120 are possible. do.

Specifically, the gas sensor 112 is installed in the safety equipment 122 of the worker 120, and the gas sensor 112 detects harmful gas generation while the worker 120 performs a certain task at the work site 110. When detected, a gas detection signal may be formed and transmitted to the server 114. Upon receiving the gas detection signal from the gas sensor 112, the server 114 may form a danger alert signal to the alarm unit of the safety equipment 122. It can be sent to (124). Upon receiving the danger warning signal from the server 114, the alarm unit 124 may notify the generation of noxious gas using sound and/or light. At this time, the server 114 may differently set the danger warning signal according to the concentration of the harmful gas. Through this, the worker 120 can recognize the harmful gas generation situation by himself even if the gas sensor 112 is not separately installed in the work site 120 performing the work, and the nearby neighbors who do not receive the corresponding danger alarm signal Other workers can also hear and / or visually recognize the alarm using sound and / or light by the alarm unit of the above-mentioned worker 120, so that the safety of the worker 120 directly exposed to danger, The worker 120 and other workers located nearby can also ensure safety.

In addition, in the case of the above-described embodiment, when the gas sensor 112 is directly installed in the safety equipment 112 of the operator 120 and detects the generation of harmful gas, the alarm unit of the safety equipment 122 is linked with the server 114. Although it is assumed that a danger warning signal is formed and transmitted in (124), it may occur when it is difficult to link with the server 114 due to the conditions of the work site 110 (for example, in a closed space or an area where communication is not possible such as a tunnel). Therefore, it is preferable to configure the gas sensor 112 to generate sound and/or light as its own alarm signal.

2 is an exemplary view showing the configuration of the server 114 according to an embodiment of the present invention.

As shown in FIG. 2 , server 114 may include one or more processors 115 , one or more memories 116 and/or transceivers 117 . As an example, at least one of these components of server 114 may be omitted, or other components may be added to server 114 . Additionally or alternatively, some of the components may be integrated and implemented, or implemented as a singular or plural entity. At least some of the components inside and outside the server 114 are connected to each other through a bus, general purpose input/output (GPIO), serial peripheral interface (SPI), or mobile industry processor interface (MIPI). Data and/or signals can be sent and received.

One or more processors 115 may control at least one component of the server 114 connected to the processor 115 by driving software (eg, instructions, programs, etc.). In addition, the processor 115 may perform operations such as various calculations, processing, data generation, and processing related to the present invention. In addition, the processor 115 may load data or the like from one or more memories 116 or store data in one or more memories 116 .

As described above, the one or more processors 115 receive the gas detection signal from the gas sensor 112 to form a danger warning signal, and transmit the formed danger warning signal to the alarm unit 124 through the transceiver 117. can do.

The one or more processors 115 receive movement path information from the safety equipment 122 through the transceiver 117 to determine the location of the worker 120 wearing the safety equipment 122 and the moving line of the worker 120. can make it

The one or more processors 115 may receive heart rate information from the safety equipment 122 through the transceiver 117 so that the manager monitors the heart rate information or breathing information of the worker 120 .

One or more processors 115 extract a pupil object by applying the face image received from the photographing unit 130 to the pre-learned neural network, and the operator 120 uses the extracted pupil object to perform the work site 110 It is determined whether the operator 120 is watching the work object, and when it is determined that the operator 120 is not watching the work object, a warning signal may be generated. According to one embodiment, one or more processors 115 may extract a pupil object from received facial images using a machine learning algorithm such as deep learning. A pupil object extraction method will be described later. Due to this configuration, the noxious gas management system 100 generates a warning signal when the worker 120 put into the work site 110 does not keep an eye on the work object or is dozing, so that the worker 120 can concentrate on the work. Additional environments may be provided.

One or more processors 115 extract the coordinates of the most protruding pupil part based on the face of the operator 120 from the extracted pupil object, generate a perpendicular line from the extracted coordinates, and When it does not intersect with the work object, it is determined that the operator 120 is not looking at the work object, and when the generated repair line intersects the work object, it is determined that the worker 120 is looking at the work object.

The one or more processors 115 acquire training face images, extract training feature point objects from the training face images, obtain labels that are body part information corresponding to the training feature point objects, and transfer the training feature point objects to a neural network. By application, training outputs corresponding to training feature point objects may be generated, and a pupil object may be extracted by learning a neural network based on the training outputs and labels.

One or more memories 116 may store various data. Data stored in the memory 116 is data obtained, processed, or used by at least one component of the server 114 and may include software (eg, commands, programs, etc.). Memory 116 may include volatile and/or non-volatile memory. In the present invention, a command or a program is software stored in the memory 116, and an operating system for controlling the resources of the server 114, an application and/or an application for various functions so that the application can utilize the resources of the server 114. It may include middleware provided to .

The one or more memories 116 may store movement path information, heart rate information, face images, and danger warning signals formed by one or more processors 115 from the plurality of safety devices 122 and the communication unit through the network. In addition, one or more memories 116 may store instructions that, when executed by one or more processors 115, cause one or more processors 115 to perform operations.

As an example, the server 114 may further include a transceiver 117 . The transceiver 117 may perform wireless or wired communication between the plurality of safety devices 122, the photographing unit 130, and/or other devices. For example, the transceiver 117 may transmit eMBB (enhanced mobile broadband), URLLC (Ultra Reliable Low-Latency Communications), MMTC (Massive Machine Type Communications), LTE (long-term evolution), LTE-A (LTE Advance), UMTS (Universal Mobile Telecommunications System), GSM (Global System for Mobile communications), CDMA (code division multiple access), WCDMA (wideband CDMA), WiBro (Wireless Broadband), WiFi (wireless fidelity), Bluetooth (Bluetooth), NFC ( Wireless communication may be performed according to a method such as near field communication (GPS), global positioning system (GPS), or global navigation satellite system (GNSS). For example, the transceiver 117 may perform wired communication according to a method such as universal serial bus (USB), high definition multimedia interface (HDMI), recommended standard 232 (RS-232), or plain old telephone service (POTS). have.

As an embodiment, one or more processors 115 may obtain information from a plurality of safety devices 122 and the photographing unit 130 by controlling the transceiver 117 . Information obtained from the plurality of safety devices 122 and the photographing unit 130 may be stored in one or more memories 116 .

As an example, server 114 may be a device of various types. For example, server 114 may be a portable communication device, a computer device, or a device according to a combination of one or more of the foregoing. The server 114 of the present invention is not limited to the devices described above.

Various embodiments of the server 114 according to the present invention may be combined with each other. Each embodiment may be combined according to the number of cases, and the embodiment of the server 114 made in combination also belongs to the scope of the present invention. In addition, internal/external components of the server 114 according to the present invention described above may be added, changed, replaced, or deleted according to embodiments. In addition, the aforementioned internal/external components of the server 114 may be implemented as hardware components.

3 is a diagram for explaining learning of a neural network according to an embodiment of the present invention.

As shown in FIG. 3 , the learning device 140 may train the neural network 142 to recognize pupils included in a face image. According to one embodiment, the learning device 140 may be a separate subject different from the server 114, but is not limited thereto.

The neural network 142 includes an input layer 141 receiving training samples and an output layer 143 outputting training outputs, and may be learned based on differences between the training outputs and labels. Here, the labels may be defined based on body part information corresponding to the feature point object. The neural network 142 is connected to a group of a plurality of nodes, and is defined by weights between the connected nodes and an activation function that activates the nodes.

The learning device 140 may train the neural network 142 using a gradient descent (GD) technique or a stochastic gradient descent (SGD) technique. The learning device 140 may use a loss function designed by the outputs and labels of the neural network 142 .

The learning device 140 may calculate a training error using a predefined loss function. The loss function may be predefined with labels, outputs and parameters as input variables, where the parameters may be set by weights in neural network 142 . For example, the loss function may be designed in a mean square error (MSE) form, an entropy form, and the like, and various techniques or methods may be employed in an embodiment in which the loss function is designed.

The learning device 140 may find weights affecting the training error by using a backpropagation technique. Here, weights are relationships between nodes in neural network 142 . The learning device 140 may use the SGD technique using labels and outputs to optimize the weights found through the backpropagation technique. For example, the learning device 140 may update weights of a loss function defined based on labels, outputs, and weights using the SGD technique.

According to an embodiment, the learning device 140 may obtain training face images and extract training feature point objects from the training face images. The learning device 140 may obtain pre-labeled information (labels) for each of the training feature point objects, and may obtain labels indicating body part information predefined in the training feature point objects.

According to an embodiment, the learning device 140 may generate training feature vectors based on appearance features, pattern features, and color features of training feature point objects. Various methods may be employed to extract features.

According to an embodiment, the learning device 140 may obtain training outputs by applying the training feature vectors to the neural network 142 . The learning device 140 may train the neural network 142 based on the training outputs and labels. The learning device 140 may train the neural network 142 by calculating training errors corresponding to training outputs and optimizing a connection relationship between nodes in the neural network 142 to minimize the training errors. The server 114 may extract a pupil object from the face image using the neural network 142 on which learning has been completed.

Figure 4 is a flow chart showing the procedure of the harmful gas management method according to an embodiment of the present invention. Although process steps, method steps, algorithms, etc. are described in a sequential order in the flowchart of FIG. 4, such processes, methods, and algorithms may be configured to operate in any suitable order. In other words, the steps of the processes, methods, and algorithms described in various embodiments of the present invention need not be performed in the order described herein. Additionally, although some steps are described as being performed asynchronously, in other embodiments some of these steps may be performed concurrently. Further, illustration of a process by depiction in the drawings does not mean that the illustrated process is exclusive of other changes and modifications thereto, and that any of the illustrated process or steps thereof may be one of various embodiments of the present invention. It does not imply that it is essential to one or more, and does not imply that the illustrated process is desirable.

As shown in FIG. 4 , in step S410 , a gas detection signal is formed when noxious gas is generated by the gas sensor 112 . For example, referring to FIGS. 1 to 3 , the gas sensor 112 installed in the work site 110 may generate a gas detection signal when noxious gas is generated.

In step S420, the gas detection signal is received by the server 114 and a danger warning signal is formed. For example, referring to FIGS. 1 to 3 , the server 114 may receive a gas detection signal from the gas sensor 112 to generate a danger warning signal.

In step S430, by the alarm unit 124, it is possible to notify the occurrence of noxious gas by using a sound and / or light by receiving a danger alarm signal. For example, referring to FIGS. 1 to 3 , the alarm unit 124 may receive a danger alert signal from the server 114 through a network and notify the generation of harmful gas using sound and/or light. According to an embodiment, the alarm unit 124 may generate an alarm signal in the form of sound and a siren in the form of light to inform the worker 120 of the noxious gas generation situation.

Although the technical idea of the present invention has been described by various embodiments above, the technical idea of the present invention includes various substitutions, modifications, and changes that can be made within the range that can be understood by those skilled in the art to which the present invention belongs. include Moreover, it is to be considered that such substitutions, modifications and alterations may be included within the scope of the appended claims.

100: harmful gas management system 110: work site
112: gas sensor 114: server
115: processor 116: memory
117: transceiver 120: operator
122: safety equipment 124: alarm unit
130: photographing unit 140: learning device
141: input layer 142: neural network
143: output layer

Claims (12)

one or more gas sensors that form a gas detection signal when noxious gas is generated;
a server configured to receive the gas detection signal and generate a danger warning signal; and
Including one or more safety equipment worn by a worker and equipped with an alarm unit for receiving the danger alarm signal and notifying the occurrence of the noxious gas using at least one of sound and light,
Hazardous Gas Management System. According to claim 1,
The one or more safety devices,
Including any one or more of safety helmets, safety shoes, safety vests, fire protection clothing, safety belts, and oxygen cylinders for fire suppression,
Hazardous Gas Management System. According to claim 1,
The one or more safety devices,
Further comprising a heart rate sensor for forming heart rate information or breathing information of the worker,
Hazardous Gas Management System. According to claim 1,
Further comprising a photographing unit for acquiring a face image of the worker working at the work site at a location spaced apart from the worker,
Hazardous Gas Management System. According to claim 4,
The server,
Applying the face image to a pre-learned neural network to extract a pupil object,
It is determined whether the worker is looking at the work object at the work site using the pupil object,
Generating a warning signal when it is determined that the operator is not looking at the work object,
Hazardous Gas Management System. According to claim 5,
The server,
Extracting the coordinates of the most protruding pupil part based on the face of the operator from the pupil object,
Generating a perpendicular line from the extracted coordinates,
When the generated perpendicular line does not intersect the work object, it is determined that the operator is not looking at the work object;
Determining that the operator is looking at the work object when the generated repair line intersects the work object,
Hazardous Gas Management System. According to claim 5,
The server,
Acquiring training face images, extracting training feature point objects from the training face images, obtaining labels corresponding to body part information corresponding to the training feature point objects, and applying the training feature point objects to the neural network, generating training outputs corresponding to training feature point objects, and extracting the pupil object by learning the neural network based on the training outputs and the labels;
Hazardous Gas Management System. According to claim 1,
The gas sensor can detect the concentration of the harmful gas,
The server forms the danger warning signal differently according to the concentration of the noxious gas,
Hazardous Gas Management System. According to claim 1,
The one or more safety devices are installed with a location recognition means capable of determining the location of the worker,
The server compares the position where the gas sensor that formed the gas detection signal is installed and the position of the worker identified through the position recognition means, selects the worker within a predetermined distance, and transmits the danger alert signal,
Hazardous Gas Management System. Forming a gas detection signal when noxious gas is generated by one or more gas sensors;
generating, by a server, a danger warning signal by receiving the gas detection signal; and
Including the step of receiving the danger alarm signal by an alarm unit provided in one or more safety equipment worn by a worker and notifying the occurrence of the noxious gas using one or more of sound and light,
Hazardous Gas Control Methods. one or more gas sensors that form a gas detection signal when noxious gas is generated;
a server configured to receive the gas detection signal and form a danger warning signal; and
An alarm unit receiving the danger warning signal and notifying the generation of the noxious gas using at least one of sound and light;
The gas sensor and the alarm unit are installed in one or more safety equipment worn by a worker,
Hazardous Gas Management System. According to claim 11,
The gas sensor further generates its own alarm signal when the harmful gas is generated,
Hazardous Gas Management System.

Images