KR102441874B1 - Fire fighting system in tunnel using CCTV - Google Patents

Abstract

In accordance with the present disclosure, in a system for suppressing a fire in a tunnel by using a CCTV, a fire detection unit (300) detects flames and smoke through a machine learning technique using image information acquired from a camera module (120) and an infrared camera (150) of a CCTV (100), thereby quickly and accurately detecting the generation of flames and smoke caused by an early fire. Also, as a fire generation position determination unit (400) accurately identifies a fire generation position, a water curtain generation unit (510) generates a water curtain in the smoke generation zone to prevent the spread of the smoke, while a tunnel upper-part fire extinguishing unit (530) and a tunnel side fire extinguishing unit (550) spray water from an upper part and a side part of the fire generation position, thereby quickly suppressing a fire. Moreover, a plurality of blower units (600) close to the fire generation position are controlled to lead the smoke generated by the fire toward the upper part and the center of the tunnel, thereby quickly exhausting the smoke caused by the early fire and preventing the spread thereof, which can lead to an effect of minimizing a loss of lives. Furthermore, the CCTV (100) can automatically clean foreign substances and effectively remove the foreign substances for better detection performance.

Description

Fire fighting system in tunnel using CCTV

The present invention relates to a system for extinguishing a fire at an early stage when a fire occurs using CCTV installed in a tunnel.

In general, as a ventilation method for road tunnels, a type type in which a jet fan hanging from the ceiling inside the tunnel forms an airflow in the direction of vehicle travel, especially in a one-way road tunnel, is the most widely used. A cross-flow ventilation method, vertical shaft ventilation method, and concentrated exhaust method, etc., are used to discharge to the outside but allow airflow to move from the road surface to the ceiling.

The jet fan type, which is the most widely used, has the disadvantage that the atmospheric environment near the exit deteriorates because all the pollutants generated by the vehicle in the tunnel are discharged to the exit, and the tunnel itself is used as a smoke exhaust passage in case of a fire.

On the other hand, in general, when a fire occurs in a tunnel, most of the casualties are caused by suffocation and reduced visibility due to toxic smoke. In other words, when a fire occurs in a tunnel, it takes a lot of time for the fire engine to enter the fire location, so it is difficult to initially extinguish the fire. It is difficult for evacuees to quickly escape from harmful gases, which can lead to many casualties.

Due to such problems, road tunnel disaster prevention facilities are being strengthened in Korea, such as making smoke control (exhaust gas) facilities compulsory for road tunnels over 500 m in length. A smoke control facility is a facility that provides a safe evacuation and evacuation environment that can minimize the impact on evacuees by controlling the direction of toxic gas or hot air generated in a fire or exhausting it from a certain area. Ventilation method using

The disaster prevention systems installed in the known tunnels are a general ventilation system (Smoke Ventilation System) and smoke ventilation equipment such as a duct and a vertical chimney (chimney) for general ventilation and smoke in case of fire. System) including a ventilation system, a tunnel air curtain for blocking the movement of smoke and toxic gas, a sprinkler and a water spraying facility for extinguishing a fire (Tunnel Sprinkler) are provided partially and in combination.

In reality, when a fire occurs in a special environment called a tunnel, the spread of heat, smoke, and toxic gas generated makes it very difficult to evacuate people and even access firefighters and rescuers for extinguishing the fire. Since it causes casualties and enormous economic loss, it is necessary to extinguish the fire early, and in particular, it is necessary to effectively control the smoke caused by the fire.

On the other hand, in general, closed-circuit television (CCTV) is installed in indoor and outdoor places such as roads, highways, and around various facilities and buildings for the purposes of traffic information collection, fire prevention, property protection, public security, and methods. have.

There are two types of CCTV cameras, one in which the camera is exposed and one in which the camera is installed inside the housing. In the case of CCTV installed outdoors, the camera is usually installed inside the housing to protect the camera.

However, CCTV installed outdoors is difficult to secure visibility as foreign substances such as rainwater, snow, and dust accumulate on the window of the housing facing the camera. If visibility is not secured, it is difficult to read the video, so the original function of the CCTV cannot be performed.

Moreover, since CCTV installed outdoors is installed in a high position that is difficult to reach by ordinary people for the purpose of installation, it is difficult to periodically clean the window if foreign substances are accumulated. In particular, since CCTV installed in the tunnel is installed at the top of the tunnel, there is a more difficult problem in cleaning or management.

Registered Patent Publication No. 10-1996633 (Registered on June 28, 2019) Registered Patent Publication No. 10-1974074 (Registered on April 24, 2019) Registered Patent Publication No. 10-2275751 (Registered on July 5, 2021) Registered Patent Publication No. 10-1885910 (Registered on July 31, 2018)

The present invention was devised in view of the above points, and when a fire occurs inside a tunnel, it is quickly detected at the beginning using CCTV to suppress the fire, and also to quickly and effectively control the flow of smoke due to the fire. The purpose of the present invention is to provide a fire suppression system in the tunnel using CCTV to discharge the fire.

Another object of the present invention is to accurately detect a foreign material attached to the window of the CCTV and also detect the attachment position of the foreign material, so that the foreign material attached to the window can be effectively removed, so that when a fire is detected using the CCTV, the CCTV This is to improve the detection performance.

A fire suppression system in a tunnel using a CCTV according to the present invention for achieving the above object includes a jet fan, a CCTV, a fire detection unit, a fire location determination unit, a water curtain generation unit, a tunnel upper fire extinguishing unit, It includes a tunnel side fire extinguishing unit, a blower unit, a wind direction and wind speed measuring unit, and a control unit.

A plurality of jet fans are installed at regular intervals in the central part of the ceiling of the tunnel and form an airflow from the entrance side to the exit side of the tunnel. The CCTV is installed in a plurality of the tunnel, the window is installed on one side, the housing body is installed to face the window and a camera module and an infrared camera for taking an image inside the tunnel are installed; a compressed air supply unit installed inside the housing body to charge and supply compressed air; a plurality of air nozzles arranged at regular intervals in a radial direction on the front side of the window to spray the compressed air supplied from the compressed air supply unit to the window; Receives an image from the camera module, and provides an object removal image, which is an image after recognizing and removing fixed installations, vehicles and other objects inside the tunnel except for the foreign substances in the received image, and analyzes the object removal image a foreign material detection unit for determining that the foreign material is attached to the window and detecting a position where the foreign material is attached when the foreign material is detected; and controlling the driving of the compressed air supply unit so that air is simultaneously sprayed from the plurality of air nozzles periodically at a set time period, and when it is determined that a foreign material is detected in the window from the foreign material detection unit, the plurality of air and a CCTV controller for controlling the driving of the compressed air supply unit so that air is sprayed from at least one air nozzle corresponding to a position where the foreign material is attached among the nozzles. The fire detection unit detects fire by detecting whether flames and smoke are generated inside the tunnel by using image information obtained from the camera module and infrared camera of the CCTV. When it is determined that the fire is detected by the fire detection unit, the fire location determination unit determines the location of the fire that has occurred inside the tunnel by using the location information of the CCTV that has detected the fire. A plurality of water curtain generators are installed at regular intervals on the upper part of the tunnel to generate a thin water curtain to block flames and smoke. A plurality of upper tunnel fire extinguishing units are installed at regular intervals so as to be installed adjacent to the water curtain generating unit at the upper portion of the tunnel, and a nozzle for spraying water, and the nozzle can be moved vertically, horizontally, or rotated to control the spraying direction. It is provided with a nozzle moving unit, and an infrared camera. A plurality of tunnel side fire extinguishing units are installed at regular intervals on the side of the tunnel, and a nozzle for spraying water, a nozzle moving part for adjusting the spraying direction by allowing the nozzle to move vertically, horizontally, or rotated, and an infrared camera. be prepared The water supply unit supplies water to the water curtain generating unit, the upper tunnel fire extinguishing unit, and the tunnel side fire extinguishing unit. A plurality of blower units are installed in the upper and side portions of the tunnel at regular intervals in the longitudinal and height directions of the tunnel, and a blower generating wind, a blower driving unit providing rotational force to the blower, and the blower It is provided with a blower moving unit that allows the air to move vertically, horizontally, or rotationally. A plurality of wind direction and wind speed measuring units are installed at regular intervals in the longitudinal direction of the tunnel to measure the wind direction and wind speed inside the tunnel. When the control unit receives the fire detection and fire location information from the fire detection unit and the fire location determination unit, the control unit controls the water curtain generation unit, the tunnel upper fire extinguishing unit, the tunnel side fire extinguishing unit, and the water supply unit to control the fire location At the same time, water curtains are generated at both ends of The wind direction of the wind generated by the blower unit and the wind direction of the wind generated by the blower unit in consideration of the wind direction and wind speed information in the tunnel received from the wind direction and wind speed measuring unit Control the wind speed.

According to the present invention, the fire detection unit detects flames and smoke using machine learning techniques using image information obtained from the camera module and infrared camera of CCTV, so that the flame and smoke generation according to the initial fire can be quickly and accurately detected. do.

In addition, it is possible to quickly extinguish the fire by accurately grasping the location of the fire and generating a water curtain in the area of the fire to prevent the spread of smoke, and at the same time spraying water from the upper and side portions of the location of the fire.

In addition, by controlling a plurality of blowers to guide the smoke generated by the fire toward the upper and middle sides of the tunnel, it is effective in minimizing the casualties by quickly evacuating the smoke from the initial fire and preventing the spread.

In addition, by analyzing the image provided by the CCCTV, the fixed installation inside the tunnel excluding foreign substances, the vehicle and other objects are recognized, and the object removal image, which is an image after removing them, is used to accurately detect the foreign material attached to the window of the CCTV and also It can detect even the attachment position of foreign substances. In addition, a plurality of air nozzles are installed radially around the window, and the air can be sprayed to the air nozzle corresponding to the position where the foreign material is attached, so that the foreign material attached to the window can be effectively removed, resulting in the detection of CCTV It has the effect of improving performance.

1 is a block diagram of a fire suppression system in a tunnel using CCTV according to an embodiment of the present invention;
2 is a schematic cross-sectional view of a tunnel to which a system according to an embodiment of the present invention is applied;
3 is an overall configuration diagram of the CCTV of FIG. 1;
Figure 4 is a schematic side cross-sectional view of the housing body of the CCTV of Figure 3;
5 is a view seen from the front of the housing body according to the example of FIG. 4;
6 is a view from the front of the housing body according to another example of FIG. 4;
7 is a block diagram of the foreign material detection unit of FIG. 3;
8 is a block diagram of the fire detection unit of FIG. 1;
9 is a schematic longitudinal cross-sectional view of a tunnel to which a system according to an embodiment of the present invention is applied;
10 and 11 are diagrams for explaining the operation of extinguishing a fire in the system of the present invention.

The above objects, features and other advantages of the present invention will become more apparent by describing preferred embodiments of the present invention in detail with reference to the accompanying drawings. Hereinafter, a fire suppression system in a tunnel using a CCTV according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

The fire suppression system according to the embodiment of the present invention is preferably applied to a one-way road tunnel, and in particular, it is to quickly extinguish a fire and to remove smoke quickly and effectively according to the occurrence of a fire.

Referring to FIG. 1 , the fire suppression system in a tunnel according to an embodiment of the present invention includes a jet fan 10 , a CCTV 100 , a fire detection unit 300 , a fire location determination unit 400 , and a water curtain generation. It includes a section 510, a tunnel upper fire extinguishing unit 530, a tunnel side fire extinguishing unit 550, a water supply unit 700, a blower unit 600, a wind direction and wind speed measuring unit 800 and a control unit 900. .

Referring to FIG. 2 , a plurality of jet fans 10 according to an embodiment of the present invention are installed at regular intervals in the center of the ceiling in the longitudinal direction of the tunnel, and the airflow from the entrance side of the tunnel T to the exit side of the tunnel T is provided. to form According to an embodiment of the present invention, the tunnel may be a type ventilation system applied.

The jet fan 10 may be increased or decreased according to the ventilation capacity or the length of the tunnel or the cross-sectional size of the tunnel, and may be usually installed every 100 to 200 m.

Referring back to FIG. 1 , the CCTV 100 is installed at a predetermined distance in the longitudinal direction in the tunnel, and includes a camera module for acquiring image information inside the tunnel. The CCTV usually provides image information for the function of the fire detection unit 300 and the fire location determination unit 400, which will be described later, as well as its original role of monitoring the situation or traffic information inside the tunnel.

There are several types of CCTVs depending on the place or purpose of installation. Usually, in the case of an outdoor, especially a tunnel, a housing-integrated CCTV having a structure in which a camera module is installed inside a housing is usually applied as in the present invention. In addition, an infrared camera capable of acquiring an infrared image in the tunnel may be installed in the CCTV in addition to the camera module.

3 and 4, the CCTV 100 according to an embodiment of the present invention is a housing body 101 having a window 110 on the front side, a camera module installed on the housing body 101 and having a monitoring function. 120, the compressed air supply unit 130 for supplying compressed air for cleaning the window 110 of the housing body 101, and the compressed air provided from the compressed air supply unit 130 is sprayed on the window 110 air nozzle 140 to remove foreign substances, an infrared camera 150 installed on the housing body to acquire an infrared image in the tunnel, and a foreign substance detection unit 160 for detecting foreign substances attached to the window 110 of the housing body 101 ) and a CCTV control unit 170 .

The housing body 101 is to protect the camera module 120 installed therein from the external environment, and a window 110 which is a transparent window is provided in front. There are several types of CCTVs depending on the place or purpose of installation. Usually, in the case of an outdoor, especially a tunnel, a housing-integrated CCTV having a structure in which a camera module is installed inside a housing is usually applied as in the present invention.

The camera module 120 is for acquiring image information and having a monitoring function, and like a normal camera module, a lens module is provided in the front and an image sensor is provided in the body. The window 110 of the housing body 101 faces the lens module of the camera module 120 . In addition, the infrared camera 150 may transmit image information acquired from the camera module 120 and other information acquired in the present system by the communication unit 180 performing a wired/wireless communication function.

The compressed air supply unit 130 supplies compressed air to wash the window 110 of the housing body 101 . Compressed air supply unit 130 may be applied to various known methods for supplying conventional compressed air, the embodiment of the present invention is not limited to a specific configuration. For example, the compressed air supply unit 130 may include a compressor for compressing air, and an air tank in which the air compressed in the compressor is temporarily stored. In addition, the compressed air supply unit 130 is connected to the air tank to provide a path through which the air of the air tank flows, and may include an air supply pipe 131 having an air valve 132 .

The air supply pipe 131 is provided with an air valve 132 to control the flow of air, and the air valve 132 is operated by the command of the CCTV controller 170 . The air valve 132 may control (ON/OFF) the flow of air flowing through the air supply pipe 131 as well as control the pressure of the compressed air. Therefore, when a foreign material is detected in the window and the foreign material remains in the window even after the operation of removing the foreign material by spraying compressed air through the air nozzle 140, the CCTV control unit 170 controls the air valve 132 to increase the pressure. It is possible to remove foreign substances that have not been removed by spraying compressed air. As will be described later, since a plurality of air nozzles 140 are installed, a plurality of the air supply pipe 131 and the air valve 132 are also installed, and the CCTV control unit 170 can independently drive the air valve 132 . can be controlled so that

The air nozzle 140 is connected to the air supply pipe 131 of the compressed air supply unit 130 to spray the compressed air supplied through the air supply pipe 131 . The air nozzle 140 sprays compressed air to remove foreign substances adhered to the window 110 of the housing body 101 . Accordingly, the air nozzle 140 is disposed near the window 110 .

Meanwhile, referring to FIGS. 5 and 6 , a plurality of air nozzles 140 may be disposed radially at regular intervals on the front side of the window 110 , and may be usually installed to be supported by the housing body 101 .

For example, when the window 110 is rectangular as shown in FIG. 5 , the air nozzle 140 may be installed one at a time on each side of the window 110 or at a position corresponding to the vicinity of the corner. It is preferable that a total of four nozzles 140 are installed, one each near the edge of the window 110 .

In addition, when the window 110 is circular as shown in FIG. 6 , a plurality of air nozzles 140 may be uniformly disposed on the circumference of the window 110 , but it is preferable that four or three are installed as shown. .

As described above, the CCTV control unit 170 can control the air valve 132 to be driven independently, so the plurality of air nozzles 140 can individually spray air, or two or more At the same time, air can be sprayed.

As shown in FIGS. 5 and 6 , when the window is divided into approximately four regions, the plurality of air nozzles 140 are radially and evenly arranged around the window, so that at least one of the four regions of the window 110 is When a foreign material is attached to one area, air can be sprayed to the air nozzle 140 corresponding to the area, and eventually, according to the present invention, the foreign material attached to the window 110 can be quickly and accurately removed. There is an advantage. If the foreign material is attached over two or more areas among the four areas, air may be sprayed to the air nozzles corresponding to each area.

Conventionally, one air nozzle is installed near the window or two or more in a line on one side to spray air uniformly. may occur, and according to the present invention, it is possible to solve this problem.

The foreign material detection unit 160 detects whether a foreign material is attached to the window surface and also determines the foreign material attachment position. Specifically, the foreign material detection unit 160 receives an image from the camera module 120 and receives the It provides an object removal image, which is an image after detecting and removing fixed installations inside the tunnel, vehicles and other objects, except for the foreign material in the image, and when a foreign material is detected by analyzing the object removal image, there is a foreign object in the window. It is determined that it is attached, and the position where the foreign material is attached is determined.

In the tunnel, it can be divided into a 'fixed installation' including a structure fixedly installed in the tunnel, for example, lighting, electric signboard, fan, and other structures, and a 'vehicle' and 'other objects' running in the tunnel. Here, 'other objects' are unusual and special cases that can be located inside the tunnel. For example, birds or insects introduced into the tunnel, information boards installed during tunnel interior construction work, workers during tunnel interior construction work, etc. can

Here, the camera module 120 acquires an image inside the tunnel through the surface of the window 110 , and the foreign material detection unit 160 includes a fixed installation excluding foreign substances in the tunnel from the image obtained from the camera module 120 . , a vehicle, other objects, etc. are recognized and the image is analyzed after removing them, and when a foreign material is detected in the image analyzed in this way, it is determined that the foreign material is detected on the surface of the window 110 .

Referring to FIG. 7 , the foreign material detection unit 160 specifically includes an image input unit 163 , an image preprocessor 163 , an object detection and removal unit 165 , and a foreign material detection and location determination unit 167 .

The image input unit 163 receives a plurality of images from the camera module 120 . The plurality of images input to the image input unit 163 may refer to, for example, a plurality of consecutive images captured by the camera module 120 for a predetermined period of time at predetermined predetermined time intervals.

The image preprocessor 163 continuously associates a plurality of captured images input to the image input unit 163 according to a time sequence, and an image required to determine a fixed installation, a vehicle, and other objects in the tunnel from the plurality of captured images. It can be pre-processed by resizing to size.

The database unit 190 (refer to FIG. 3) stores image data input through the image input unit 163, and also stores various types of image data acquired through a path other than the image input unit. Images and outline data, images and outline data of the various vehicles (cars, trucks, etc.), and images and outline data of the other objects may be stored. The collected data is used as machine learning-based training data and test data. In addition, the database unit 190 includes an operating program of the system, a plurality of threshold information to be used in image analysis of the foreign material detection and location determination unit 167 to be described later, and determination generated by the foreign material detection and location determination unit 167 . Information and the like may be stored.

The object detection and removal unit 165 recognizes and extracts fixed installations, vehicles and other objects in the tunnel from the pre-processed image resizing by the image preprocessor 163, and extracts the extracted tunnel from the pre-processed image. The fixed installation, vehicle, and other objects inside are removed and provided to the foreign material detection and location determination unit 167 to be described later.

Various methods may be applied to a method for the object detection and removal unit 165 to recognize a fixed installation, a vehicle, and other objects.

First, in the case of fixed installations, outlines of fixed installations may be detected by edge detection in the pre-processed image. Here, since the image pre-processing unit 163 pre-processes a plurality of images continuously input from the image input unit 163, the fixed installation in the plurality of images is fixed without movement or movement, and the contours detected in each image are also It will be in the same position in multiple images. Accordingly, when the outline of the extracted object is at the same location in the plurality of images, the object detection and removal unit 165 recognizes it as a fixed installation.

On the other hand, in the case of a vehicle, its position is moved in a plurality of images, unlike a fixed installation. Accordingly, the contour of the vehicle object is detected by edge detection in the pre-processed image, and when the contour of the vehicle object has different positions in a plurality of images, it is recognized as a vehicle.

And in the case of other objects, for example, birds, bugs, or workers, as they move over time like the aforementioned vehicle objects, they can be recognized by processing them in the same process as the vehicle objects, for example, during tunnel interior construction work. In the case of a guide plate to be installed, since it is fixed, it can be recognized by processing in the same process as the above-described fixed installation.

On the other hand, the object detection and removal unit 165 uses various contour data stored in the database unit 190 as learning data to complete the learning and analyze it based on an image artificial intelligence algorithm to analyze the fixed installations, vehicles and other objects. can be recognized and detected. Here, the image artificial intelligence algorithm is, for example, a convolutional neural network (CNN, convolutional neural network), a recurrent neural network (RNN), a deep neural network, etc. A neural network model can be applied.

The foreign material detection and location determination unit 167 is a foreign material attached to the surface of the window 110 from the fixed installation provided by the object detection and removal unit 165, the vehicle and other objects are removed from the image (hereinafter 'object removal image'). It detects whether there is a problem and also detects the position of the foreign material.

As described above, the object detection and removal unit 165 recognizes, extracts, and removes fixed installations, vehicles, and other objects in the tunnel. Since is removed, only the background material such as the ceiling or wall remains, and it is provided to the foreign material detection and position determination unit 167 . Accordingly, the foreign material detection and location determination unit 167 detects other things than the background such as a ceiling or a wall in the 'object removal image' provided by the object detection and removal unit 165, it detects that the foreign material is attached to the surface of the window 110. will be judged to be

The foreign material detection and location determination unit 167 calculates a brightness value for each image by using the first inputted object removal image among the object removal images. Here, the calculated brightness value (hereinafter, 'calculated brightness value') is information used to determine a state in which foreign substances are already attached to the surface of the window 110. By comparing the 'calculated brightness value' with the 'reference brightness value', When the system starts an operation for determining the detection of a foreign material, it may be checked whether a foreign material is attached to the surface of the window 110 before a plurality of images are input to the image input unit 163 . In this case, by comparing the calculated brightness value with the reference brightness value, it can be checked whether foreign substances are attached to the surface of the window 110 from the time the system starts operating. If it is confirmed that the foreign material is already attached to the surface of the window 110 from the time the system starts operating, which area of the window 110 is the foreign material attached to the first input image among the plurality of images as the target image? will judge

Here, the 'reference brightness value', which is the brightness value before the foreign material is attached to the surface of the window 110 , is stored in the database unit 190 . This 'reference brightness value' is measured in a state in which there is no foreign substance on the surface of the window 110 when the system of the present invention is installed and applied, or the brightness in a state in which there is no foreign substance is periodically measured and stored in the database unit 190. can be saved.

However, if it is confirmed that the foreign material is not attached to the surface of the window 110 at the time when the foreign material detection operation of the system starts, it is determined on which area of the window 110 surface the foreign material is attached using the subsequently input image as the target image. will judge

Using the image data designated as the target image among the object removal images, the foreign substance detection and position determination unit 167 sequentially lists the brightness values of each pixel of the image data to generate a Bezier curve. The Bézier curve is generated by the Bézier algorithm, and the Bézier algorithm is to walk the n-1st-order curve from n points, and there are representative mid-point methods and B-spline methods.

Thereafter, the foreign material detection and position determination unit 167 detects a pixel in which the vertical distance between the difference value of each pixel and the Bezier curve is equal to or greater than a predetermined threshold value, and the detected pixel greater than or equal to the threshold value is attached to the surface of the window 110 . Determination information on whether or not it has been determined is generated.

On the other hand, when it is determined that the foreign material is attached to the window, the foreign material detection and position determination unit 167 also detects a location where the foreign material is attached. That is, the foreign material detection and location determination unit 167 is able to know the position of the corresponding pixel of the camera module 120 to which the foreign material is attached, and transmits the foreign material detection determination and the foreign material attachment location information to the CCTV control unit 170 . .

The CCTV control unit 170 controls the operation of the compressed air supply unit and the air nozzle so that compressed air is periodically supplied to the window 110 for a set period of time, thereby removing dust accumulated on the window surface.

The type of foreign material may be divided into dust accumulated on the surface of the window over time and specific foreign matter other than the dust. The specific foreign material is, for example, small insects or worms attached to the window surface, or birds or insects. or insect excrement, or other foreign substances.

As such, the CCTV control unit 170 periodically supplies compressed air to the window 110 , thereby removing dust accumulated on the surface of the window 110 for a certain period of time. In this case, since the CCTV control unit 170 needs to clean the entire surface of the window, the compressed air is sprayed from all four air nozzles 140 shown in FIGS. 5 and 6 .

On the other hand, the specific foreign material other than the dust is attached to the window 110, and the foreign material (meaning the specific foreign material other than the dust) is detected in the window 110 from the foreign material detection unit 160, and the location information of the foreign material When receiving, the compressed air supply unit 130 is driven so that the air of the air nozzle 140 corresponding to the area to which the foreign material is attached is sprayed, so that the compressed air is sprayed into the air nozzle 140 .

For example, in FIGS. 5 and 6 , the foreign material detection and position determination unit 167 of the foreign material detection unit 160 determines that a foreign material is detected in the upper right region of the window 110 and transmits this information to the control unit. , CCTV control unit 170 drives the compressed air supply unit 130 so that compressed air is injected from the air nozzle 140 corresponding to the upper right area. On the other hand, for example, when it is determined that the foreign material is attached to overlap the plurality of areas, the CCTV control unit 170 may cause compressed air to be sprayed from the plurality of air nozzles 140 corresponding to the plurality of areas.

As described above, according to the present invention, in the case of foreign substances such as dust, it is possible to remove the dust by periodically spraying compressed air on the surface of the window. And in the case of foreign substances other than dust, the image provided by the CCCTV is analyzed to recognize fixed installations, vehicles and other objects inside the tunnel except for the foreign substances, and the object removal image is an image after removing the foreign substances attached to the window of the CCTV. Accurately detects and also detects the attachment location of foreign substances. In addition, a plurality of air nozzles are radially installed around the window, and air can be sprayed to the air nozzles corresponding to the positions where the foreign substances are attached, so that there is an effect that can effectively remove the foreign substances attached to the window.

Referring back to FIG. 1 , the fire detection unit 300 uses the image information obtained from the camera module 120 and the infrared camera 150 of the CCTV 100 to detect a fire inside the tunnel, that is, flames and smoke. It detects whether or not a detection has occurred, and transmits a detection signal to the control unit 900 and the fire occurrence location determination unit 400 . Referring to FIG. 8 , the fire detection unit 300 specifically includes an image input unit 310 , an image preprocessor 330 , and a flame and smoke detection unit 350 .

The image input unit 310 receives a plurality of images from the camera module 120 of the CCTV 100 . The plurality of images input to the image input unit 310 may refer to, for example, a plurality of consecutive images captured by the camera module for a predetermined period of time at predetermined predetermined time intervals. Also, the image input unit 310 receives a plurality of images from the infrared camera 150 .

The image preprocessor 330 continuously associates a plurality of captured images input to the image input unit 310 according to a time sequence, and resizes the image size required to discriminate flames and smoke from the plurality of captured images to preprocess the image. can do.

The image data input through the image input unit 310 is stored in the not-shown database unit, and various types of image data acquired through a path other than the image input unit are stored. In particular, various vehicles (cars, trucks, etc.) image and contour data of, image (size, shape) and contour data of a flame or flame according to a fire, smoke image and contour data according to a fire occurrence, temperature and frequency data according to a flame or flame, i.e. , Images and outline data for flames or flames and smoke according to the occurrence of a fake fire can be stored. The collected data is used as machine learning-based training data and test data. In addition, an operation program of the system may be stored in a database unit (not shown).

The flame and smoke detection unit 350 detects a flame or flame and smoke according to a fire with respect to the image pre-processed for resizing by the image pre-processing unit 330 . The smoke detection unit 1350 uses various data stored in the unillustrated database unit as learning data to complete learning and analyzes it based on an image artificial intelligence algorithm to analyze the vehicle and the flame (flame) and smoke generated from the vehicle. can be recognized and detected. Here, the image artificial intelligence algorithm is, for example, a convolutional neural network (CNN, convolutional neural network), a recurrent neural network (RNN), a deep neural network, etc. A neural network model can be applied.

In this way, the fire detection unit 300 detects the occurrence of flames and smoke according to the actual fire generated in the vehicle rather than due to the fake fire by the image artificial intelligence algorithm, thereby increasing speed and accuracy.

Thereafter, when it is determined that smoke is detected, the smoke detector 350 also detects a smoke generating area. That is, the smoke detection unit 350 can know the position of the corresponding pixel of the camera module in which the smoke is generated, and transmits the smoke generation position information to the control unit 900 .

In this way, the fire detection unit 300 uses a machine learning algorithm from the image information obtained from the camera module 120 and the infrared camera 150 of the CCTV 100 to quickly and accurately detect the flame and smoke generation according to the initial fire. be able to do

Referring back to FIG. 1 , when it is determined that the fire is detected by the fire detection unit 300 , the fire location determination unit 400 determines the location where the fire occurred and the location where the smoke flows due to the fire. be able to That is, when it is determined that a fire is detected by the fire detection unit 300 as described above, a flame and smoke detection signal due to a fire and location information of a smoke generating area on the camera module are transmitted to the control unit 900, and the control unit 900 900 receives the location information of the CCTV 100 that has detected the fire and also receives the location information of the fire area on the camera module of the CCTV to determine the location of the fire in the tunnel. The control unit 900 may rotate the CCTV to accurately track the location of the fire.

The fire occurrence location determination unit 400 may store location information of the CCTV 100, 3D map information in the tunnel, and the like. It can be expressed on a mapped three-dimensional map, and the three-dimensional map can have coordinates on the x, y, and z axes, so that it is possible to determine at which location a fire occurs.

The water curtain generator 510 is installed at regular intervals in the upper part of the tunnel to generate a thin water curtain to prevent smoke and toxic gas from expanding. The water curtain generating unit 510 supports a plurality of nozzles for spraying water by connecting a water supply unit 700 and a pipe to be described later, and supports the plurality of nozzles, and also generates a thin-film water curtain using water sprayed from the nozzles. It may include a nozzle support to make it happen. The structure of the nozzle support may be adopted in various known ways. For example, the nozzle support has a box body in which a plurality of nozzles are supported therein and the lower surface is open, and the nozzle is directed upward or forward. to eject water, and a blocking film formed in the height direction on the surface opposite to the nozzle may be installed. Accordingly, the water sprayed from the nozzle flows down to the lower part of the open nozzle support while colliding with the blocking film, thereby generating a thin water curtain.

According to this embodiment, as shown in FIG. 2 , the water curtain generating units 510 are preferably disposed at intervals of 10 to 30 m. When the control unit 900 receives the smoke generation location information from the fire occurrence location determination unit 400, the control unit 900 operates the water curtain generation unit 510 at the corresponding position so that water curtains are formed at both ends of the smoke generation location. .

The upper tunnel fire extinguishing unit 530 is installed adjacent to the water curtain generating unit 510 at the upper part of the tunnel, and sprays water to the lower part, that is, the location of the fire. The upper tunnel fire extinguishing units 530 are installed at regular intervals in the longitudinal direction of the tunnel. Specifically, referring to FIGS. 9 to 11 , the upper tunnel fire extinguishing unit 530 includes a nozzle 532 , a nozzle moving unit 534 , and an infrared camera.

The nozzle 532 receives water from the water supply unit 700 and sprays water. The nozzle moving unit 534 is installed to be supported by the nozzle body (not shown), and it is possible to adjust the spraying direction of the nozzle in up, down, left, and right directions under the control of the control unit 900 . The adjustment method and driving principle of the nozzle moving unit 534 are operated to enable vertical, horizontal, and 360-degree movement or 360-degree rotation with the aforementioned various structures, for example, a cam, a keyer, a crank structure, or a combination thereof. and manual direction adjustment may be possible. An infrared camera (not shown) may detect the location of the fire source by detecting the temperature of the flame at the location of the fire, and transmits the detected image to the controller 900 .

A plurality of tunnel side fire extinguishing units 550 are installed at regular intervals on the side of the tunnel and spray water to a location where a fire occurred on the side, that is, a road in the tunnel. As shown in FIG. 9 , the tunnel side fire extinguishing unit 550 is preferably installed at a slightly higher position than the vehicle. Specifically, referring to FIGS. 9 to 11 , the tunnel side fire extinguishing unit 550 includes a nozzle 552 , a nozzle moving unit 554 , and an infrared camera (not shown).

The nozzle 552 receives water from the water supply unit 700 and sprays water. The nozzle moving unit 554 is installed to be supported by the nozzle body, and it is possible to adjust the spraying direction of the nozzle in up, down, left, and right directions under the control of the control unit 900 . The adjustment method and driving principle of the nozzle moving unit 554 is operated to enable vertical, horizontal, horizontal, or 360-degree rotation with the aforementioned various known structures, for example, a cam, a keyer, a crank structure, or a combination thereof, and also automatically and manual direction adjustment may be possible. An infrared camera (not shown) may detect the location of the fire source by detecting the temperature of the flame at the location of the fire, and transmits the detected image to the controller 900 .

As described above, according to the present invention, the control unit 900 adjusts the positions of the nozzles of the upper tunnel fire extinguishing unit 530 and the tunnel side fire extinguishing unit 550 over two steps to spray water.

First, when receiving the fire detection and fire location information from the fire detection unit 300 and the fire location determination unit 400, the control unit 900 controls at least one upper tunnel fire extinguishing unit close to the location where the fire occurred. By controlling the nozzle moving part 534 of 530 and the nozzle moving part 554 of the tunnel side fire extinguishing unit 550, the nozzles 532 and 552 are directed to the approximate location where the fire occurred. After that, when a detection image is received from each of the infrared cameras of the tunnel upper fire extinguishing unit 530 and the tunnel side fire extinguishing unit 550, the detected image is analyzed so that the position of the flower source is located approximately in the center of the image. After controlling the positions of the moving parts 534 and 554 to make detailed adjustments so that each of the nozzles 532 and 552 are directed toward the flower garden, water can be sprayed to the nozzles through the water supply part 700 as the final group. there will be

The water supply unit 700 supplies water to the water curtain generator 510, the tunnel upper fire extinguishing unit 530, and the tunnel side fire extinguishing unit 550, for example, a water tank for storing water, a water tank It may include a pump for supplying water from have.

A plurality of blower units 600 are installed on the upper and side portions of the inside of the tunnel, and are installed at regular intervals in the longitudinal direction and the height direction of the inner wall of the tunnel. Here, the interior of the tunnel has a hemispherical structure, and the upper part of the interior of the tunnel includes the ceiling (the highest point inside the tunnel) and means a place close to the ceiling, and the side of the tunnel is a certain distance from the ceiling and can be interpreted as meaning a place close to the ground. The blower unit 600 blows wind into the tunnel interior space under the control of the CCTV controller 170 to be described later.

Referring to FIG. 9 , the blower unit 600 includes a blower 610 for generating wind, a blower driving unit (not shown) such as a motor providing rotational force to the blower 610 , and a blower 610 . ) is provided with a blower moving unit 630 to rotate and move up, down, left and right.

The control unit 900 may control the wind speed of the blower 610 by adjusting the motor rotation speed of the blower driving unit, and control the blower moving unit 630 to control the wind direction of the blower 610 . do. That is, the blower moving unit 630 is installed to support the blower 610 and is installed so as to be able to move vertically or horizontally or rotationally, and eventually the blower 610 is also rotated vertically, horizontally, or rotated by the blower moving unit 630 . It works so that it can move. The blower moving unit 630 is operated to enable vertical and horizontal flow or 360-degree rotation with a variety of known structures, for example, a cam, keyer, crank structure, or a combination thereof, and can also be adjusted automatically and manually. can The blower moving unit 630 controls the movement of the blower 610 according to the control of the controller 900 , so that it is possible to eventually control the wind direction coming out of the blower 610 .

Referring back to FIG. 1 , the wind direction and wind speed measuring unit 800 is a device for evaluating and measuring air movement inside the tunnel, and may be a device capable of measuring not only the direction of the wind but also the strength of the wind. have. Therefore, the wind direction and wind speed measurement unit 800 measures the strength and direction of the wind and transmits the measurement information to the control unit 900, and the control unit 900 uses the measurement information to predict the direction of smoke according to the fire It can help guide movement effectively.

The control unit 900 performs an overall control operation of each component of the present system. That is, the control unit 900 normally controls the wind speed of the jet fan 10 so that ventilation in the tunnel is smooth.

On the other hand, when the control unit 900 receives the fire detection and fire location information from the fire detection unit 300 and the fire location determination unit 400 as described above, the water curtain generator close to the fire location 510, the upper tunnel fire extinguishing unit 530, the tunnel side fire extinguishing unit 550, and the water supply unit 700 are controlled to generate water curtains at both ends of the area where smoke is generated, and at the same time a fire occurs. Water spray at the location to extinguish the fire early and prevent the spread of smoke and toxic gases to other areas.

Then, the control unit 900 controls the movement of at least one blower 610 and the blower moving unit 630 installed in the area where the smoke is generated, so that the smoke generated at the smoke generating position is directed toward the center of the tunnel ceiling. Controlled to induce flow, eventually causing the smoke to move towards the tunnel exit by the airflow of the jet fan (10). That is, in order to increase the smoke effect, the jet fan 10 is guided toward the top and center of the installed ceiling, and the flow is controlled to induce the flow in the forward direction of the jet fan.

Due to the characteristics of the ventilation facilities and the fire generated inside the tunnel, the smoke rises upward to some extent, but due to the nature of the sealed structure inside the tunnel, the smoke spreads not only upwards but also to the side, and serious human damage is caused by this smoke. will do Therefore, according to the present invention, the control unit 900 can control the blower 610 and the blower moving unit 630 to induce the smoke generated by the fire to flow toward the ceiling more quickly, In the event of a fire, the loss of life due to smoke can be minimized.

Hereinafter, with reference to FIGS. 10 and 11 , an operation of the fire suppression system in a tunnel according to an embodiment of the present invention will be described.

In normal times when a fire does not occur, the jet fan 10 ventilates the tunnel, and the wind volume or speed may change according to the traffic volume in the tunnel.

Next, an operation in case of fire will be described.

The fire detection unit 300 detects that flames and smoke are generated in the tunnel by using the information received from the CCTV 100 . When the fire detection unit 300 finally determines that it has detected the flame and smoke according to the fire, the fire detection unit 300 transmits the fire detection signal and the location information of the fire area on the camera module of the CCTV to the control unit 900, , the fire location determination unit 400 receives the information from the control unit 900 to determine the location of the fire.

When receiving flame and smoke detection and generation location information according to a fire from the fire detection unit 300 and the fire location determination unit 400, the control unit 900 controls the water curtain generation unit 510 close to the location where the fire occurred. , by controlling the upper tunnel fire extinguishing unit 530, the tunnel side fire extinguishing unit 550, and the water supply unit 700 to generate water curtains (dotted lines in FIG. Water jets (indicated by solid lines in FIGS. 10 and 11) from the top and sides of the location are used to extinguish the fire and prevent the spread of smoke and toxic gases to other areas.

At this time, as described above, since the upper tunnel fire extinguishing unit 530 and the tunnel side fire extinguishing unit 550 are equipped with infrared cameras, respectively, to move in search of a fire source, fire suppression can be performed more quickly. That is, firstly, the control unit 900 gives a control command to spray water to one or more tunnel upper fire extinguishing units 530 and tunnel side fire extinguishing units 550 close to the location of the fire, and the corresponding tunnel upper fire extinguishing unit 530 . ) and the tunnel side fire extinguishing unit 550 each have an infrared camera, so they find a fire source at the location of the fire and move up, down, left and right or rotationally to spray water in the direction of the fire source.

On the other hand, the control unit 900 controls the movement of at least one blower 610 and the blower moving unit 630 installed in the area where the smoke is generated, so that the smoke generated at the smoke generating position is transferred to the center of the tunnel ceiling. to control the flow to the side. That is, the blower 610 is installed in the height direction of the tunnel wall, so that all the smoke in the lower part of the tunnel, the middle of the tunnel, and the upper part of the tunnel is removed from the upper part of the tunnel where the jet fan 10 is installed (based on the height direction) and the center of the tunnel ( width direction) to guide it quickly.

In this case, the control unit 900 may receive information on the strength and direction of the wind from the wind direction and wind speed measurement unit 800 and use the information to predict the direction of fire and smoke to effectively guide the movement of smoke. That is, when the control unit 900 guides the movement of the smoke. Using the wind direction and wind speed information in the tunnel, the wind speed and the wind direction of the blower 610 are controlled (indicated by dotted lines in FIGS. 10 and 11 ), thereby induced to accurately concentrate exhaust smoke toward the upper part and the center of the tunnel.

On the other hand, although smoke is generated at the location of the fire, the generated smoke may spread not only in the upper part but also in the longitudinal direction of the tunnel. By controlling the blower corresponding to the smoke spread in the direction, it is possible to direct the smoke flow towards the tunnel ceiling and the center.

As described above, according to the present invention, by using the image information obtained from the camera module of the CCTV by the fire detection unit to detect the flame and smoke using the machine learning technique, it is possible to quickly and accurately detect the flame and smoke generation according to the initial fire. In addition, it is possible to quickly extinguish the fire by identifying the location of the fire and generating a water curtain in the area of the fire to prevent the spread of smoke, and at the same time spray water from the top and side of the location of the fire. By controlling the smoke and guiding the smoke generated by the fire toward the upper and middle of the tunnel, there is an advantage in that the smoke from the initial fire is quickly exhausted and the spread is prevented, thereby minimizing human casualties.

In addition, it is possible to accurately detect a foreign material attached to the window of the CCTV and also detect the attachment position of the foreign material, and a plurality of air nozzles are installed radially around the window, and air is As it can be sprayed, it is possible to effectively remove foreign substances attached to the window, thereby increasing the detection performance of CCTV.

Although preferred embodiments of the present invention have been described above, the present invention is not limited to the specific embodiments described above. That is, a person of ordinary skill in the art to which the present invention pertains can make numerous changes and modifications to the present invention without departing from the spirit and scope of the appended claims, and all such appropriate changes and modifications are possible. Equivalents are to be considered as falling within the scope of the present invention.

10. Jetfan 100. CCTV
101. Housing body 110. Window
120. Camera module 140. Air nozzle
150. Infrared camera 160. Foreign matter detection unit
300. Fire detection unit 400. Fire location determination unit
510. Water curtain generator 530. Tunnel upper fire extinguishing unit
550. Tunnel side fire extinguishing unit 600. Blower unit
800. Wind direction and wind speed measuring unit 900. Control unit

Claims (4)

a plurality of jet fans installed at regular intervals in the central part of the ceiling of the tunnel and forming an airflow from the entrance side to the exit side of the tunnel;
a housing body installed in a plurality of tunnels, a window is installed on one side, and a camera module and an infrared camera installed facing the window to take an image inside the tunnel; a compressed air supply unit installed inside the housing body to charge and supply compressed air; a plurality of air nozzles arranged at regular intervals in a radial direction on the front side of the window to spray the compressed air supplied from the compressed air supply unit to the window; Receives an image from the camera module, and provides an object removal image, which is an image after recognizing and removing fixed installations, vehicles and other objects inside the tunnel excluding foreign substances from the received image, and analyzing the object removal image a foreign material detection unit which determines that the foreign material is attached to the window and detects a position where the foreign material is attached when the foreign material is detected; and controlling the driving of the compressed air supply unit so that air is simultaneously sprayed from the plurality of air nozzles periodically at a set time period, and when it is determined that a foreign material is detected in the window from the foreign material detection unit, the plurality of air CCTV including a;
a fire detection unit for detecting fire by detecting whether flames and smoke are generated inside the tunnel by using the image information obtained from the camera module and infrared camera of the CCTV;
When it is determined that the fire is detected by the fire detection unit, a fire location determination unit for determining the location of a fire that has occurred inside the tunnel by using the location information of the CCTV that has detected the fire;
a water curtain generator installed at regular intervals on the upper part of the tunnel to generate a thin-film water curtain to block flames and smoke;
A plurality of nozzles are installed at regular intervals so as to be installed adjacent to the water curtain generating unit at the upper portion of the tunnel, and a nozzle moving unit configured to control the injection direction by allowing the nozzle to move vertically, horizontally, or rotated. and a tunnel upper fire extinguishing unit having an infrared camera;
Tunnel side fire extinguishing having a plurality of nozzles installed at regular intervals on the side of the tunnel and spraying water, a nozzle moving part for adjusting the spraying direction by enabling the nozzle to move vertically, horizontally, or rotated, and an infrared camera unit;
a water supply unit for supplying water to the water curtain generating unit, the tunnel upper fire extinguishing unit, and the tunnel side fire extinguishing unit;
A plurality of blowers installed at regular intervals in the longitudinal and height directions of the tunnel in the upper and side portions of the tunnel, a blower generating wind, a blower driving unit providing rotational force to the blower, and the blowers vertically and horizontally Or a blower unit having a blower moving unit to enable rotational movement;
a plurality of wind direction and wind speed measuring units installed at regular intervals in the longitudinal direction of the tunnel to measure the wind direction and wind speed inside the tunnel; and,
When the fire detection and fire location information are received from the fire detection unit and the fire location determination unit, the water curtain generation unit, the tunnel upper fire extinguishing unit, the tunnel side fire extinguishing unit, and the water supply unit are controlled to control the amount of the fire location At the same time as generating a water curtain at the end, water is sprayed at the location of the fire, and by controlling at least one blower unit installed in the vicinity of the location of the fire, smoke according to the fire is transferred to the upper part of the ceiling of the tunnel. and to guide the flow in the forward direction of the airflow of the jet fan, and to guide the wind direction and wind speed of the wind generated by the blower unit in consideration of the wind direction and wind speed information in the tunnel received from the wind direction and wind speed measuring unit A fire suppression system in a tunnel using CCTV, comprising: a control unit to control.
The method of claim 1,
The control unit is
First, when receiving the fire detection and fire location information from the fire detection unit and the fire location determination unit, at least one or more tunnel upper fire extinguishing units close to the location of the fire and each nozzle moving unit of the tunnel side fire extinguishing unit are controlled to direct each nozzle to the location where the fire occurred,
After that, when receiving a detection image from each of the infrared cameras of the fire extinguishing unit at the upper part of the tunnel and the fire extinguishing unit at the side of the tunnel, the detected image is analyzed and the respective nozzle moving parts are controlled so that the position of the flower source is located in the central portion of the image to the flower garden. A fire suppression system in a tunnel using CCTV, characterized in that after the nozzles are directed to each other, water is sprayed to the nozzles through the water supply unit.
The method of claim 1,
The fire detection unit,
an image input unit for receiving a plurality of images from the camera module of the CCTV and the infrared camera;
an image pre-processing unit for resizing and pre-processing an image size required to discriminate a vehicle and smoke in a tunnel from a plurality of captured images and infrared images input to the image input unit; and,
When it is finally determined that the flame and smoke according to the fire are detected after recognizing and extracting the flame and smoke according to the fire in the tunnel by analyzing the image pre-processed for resizing by the image pre-processing unit based on the image artificial intelligence algorithm, A flame and smoke detection unit that transmits a smoke detection signal and location information of a smoke generation area on the camera module of the CCTV to the control unit;
Tunnel using CCTV, characterized in that the fire occurrence location determination unit determines the location of the fire in the tunnel by using the location information of the CCTV that has detected the fire from the controller and the location information of the fire area on the camera module of the CCTV. My fire suppression system.
The method of claim 1,
The foreign material detection unit of the CCTV,
an image input unit receiving a plurality of images from the camera module;
an image pre-processing unit for resizing and pre-processing an image size required to identify fixed installations, vehicles, and other objects in the tunnel from the plurality of captured images input to the image input unit;
Objects obtained by recognizing and extracting fixed installations, vehicles, and other objects in the tunnel from the image preprocessed for resizing by the image preprocessor, and removing fixed installations, vehicles, and other objects in the extracted tunnel from the preprocessed image an object detection and removal unit providing a removal image; and,
A foreign material detection and location determination unit that determines whether a foreign material is attached to the window surface from the object removal image provided by the object detection and removal unit, and also detects the location of the foreign material; Fire suppression system in tunnel.

Images