KR20150078049A - Apparatus for sensing fire - Google Patents

Abstract

Disclosed is a fire detection apparatus capable of preventing false fire alarm by judging fire outbreak accurately. The disclosed fire detection apparatus includes: a real imaging camera which generates image data by photographing real images, and detects smoke; a thermal imaging camera which generates image data by photographing thermal images, and detects temperature; a flame detector for detecting flame; a data analysis and processing part which collects the image data photographed by the real imaging camera and the thermal imaging camera, and analyzes the collected data, and judges fire outbreak in case at least two of smoke, high heat and flame are generated in a monitoring target area; an output part for receiving an image signal from the data analysis and processing part and then outputting the image signal on a screen; and an alarm part which executes warning if the data analysis and processing part judges fire outbreak. By this fire detection apparatus, the present invention can obtain the effect of assuring reliability in fire monitoring and warning, by analyzing smoke, heat and flame overall.

Description

{APPARATUS FOR SENSING FIRE}

FIELD OF THE INVENTION The present invention relates to a fire detection apparatus, and more particularly, to a fire detection apparatus capable of precisely determining whether or not a fire is present, thereby preventing a fire alarm.

As the industry develops, the types of fire occur, the size of the damage increases, and fire prediction and monitoring are becoming increasingly difficult.

Various types of fire detection devices are installed, but due to aging or lack of adaptability to the surrounding environment, frequent fire alarms occur, and the resulting socio-economic losses are increasing.

For example, in a conventional fire detection system in which a fire is detected by capturing a thermal image, there is a problem in that the fire detection is performed by sensing the temperature of the human body or the temperature of the lighting apparatus as heat.

In addition, there is a problem that it is difficult to identify whether or not smoke is caused by fire only by using a real image.

Disclosure of Invention Technical Problem [8] Accordingly, the present invention has been made to solve at least some of the above-mentioned problems occurring in the prior art, and an object of the present invention is to provide a fire detection device capable of accurately detecting fire by analyzing and judging fumes, The purpose.

According to one aspect of the present invention for achieving at least part of the above objects, there is provided an image processing apparatus comprising: a real image camera for photographing a real image to generate image data; An infrared camera for capturing a thermal image to generate image data and sensing temperature; A flame detector for detecting flame; The image data photographed by the real image camera and the thermal image camera and the detection signal of the flame sensor are collected and analyzed and data which is judged as a fire occurrence when at least two of mist, An analysis processing unit; An output unit receiving the video signal from the data analysis processing unit and outputting the video signal to the screen; And an alarm unit for executing a warning when the data analysis processing unit determines that a fire has occurred.

In one embodiment, the real image camera may be configured to generate photographed image data when motion is detected in an area to be photographed.

Also, in one embodiment, the data analysis processing unit may analyze the image data taken by the real camera to detect the size of the mist, compare the detected size of the mist with a preset value, have.

Also, in one embodiment, the data analysis processing unit may output a haze image to the output unit and operate the alarm unit when a fire caused by a fire is detected and a detection signal of the flame sensor is received.

Also, in one embodiment, the data analysis processing unit may detect the magnitude of fog by performing brightness contrast adjustment, noise removal, shape-based processing, threshold value comparison noise removal, and boundary detection processing on the image data photographed by the real image camera have.

Also, in one embodiment, the data analysis processing unit may erode and expand the image in the shape-based processing.

Also, in one embodiment, the data analysis processing section analyzes the image photographed by the real image camera and the image photographed by the thermal imaging camera to extract a matching area between the real image and the thermal image, (Picture In Picture).

Further, in one embodiment, the data analysis processing unit may output the matching area as an image to the output unit and operate the alarm unit when the matching area is extracted.

Also, in one embodiment, the data analysis processing unit may perform brightness contrast adjustment, noise reduction, shape-based processing, threshold value comparison noise removal, and boundary detection processing on the image photographed by the real image camera and the image photographed by the thermal image camera A matching area between the real image and the thermal image can be extracted.

In one embodiment, the data analysis processing unit analyzes the image captured by the thermal imaging camera to detect the size of the detection target, compares the detected size of the detection target with the minimum setting size, Can be compared with a predetermined time to determine whether the fire is caused by a high temperature.

Also, in one embodiment, the data analysis processing unit may output a thermal image to the output unit and operate the alarm unit when a high temperature due to fire is detected and a detection signal of the flame sensor is received.

Further, in one embodiment, the data analysis processing section may perform a color map application, a maximum temperature and a minimum temperature extraction, a detection region setting, a detection range temperature setting, and a temperature binning setting on the image captured by the thermal imaging camera, Can be detected.

According to one embodiment of the present invention having such a configuration, the effect of ensuring reliability in fire monitoring and alarm can be obtained by comprehensively analyzing mist, heat, and flame.

1 is a block diagram showing a configuration of a fire detection apparatus according to an embodiment of the present invention;
2 is a flowchart showing a method of analyzing a real image of a data analysis processing unit included in a fire detection apparatus according to an embodiment of the present invention.
3 is an illustrative photograph for explaining a method of adjusting brightness and contrast of a video image by a data analysis processing unit;
4 is an illustrative photograph for explaining a result of eliminating the image noise of the data analysis processing unit;
5 is an illustrative photograph for explaining an image shape-based processing method of the data analysis processing unit.
FIG. 6 is an illustrative photograph for explaining an image-based processing result of the data analysis processing unit. FIG.
7 is an illustration for explaining the result of the video threshold value comparison noise removal processing of the data analysis processing unit;
8 is an illustrative photograph for explaining an image boundary detection processing result of the data analysis processing unit;
9 is a flowchart showing a method of analyzing a real image and a thermal image of a data analysis processing unit included in a fire detection apparatus according to an embodiment of the present invention.
10 is a flowchart showing a thermal image analysis method of a data analysis processing unit included in a fire detection apparatus according to an embodiment of the present invention.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. Furthermore, the singular forms "a", "an," and "the" include plural referents unless the context clearly dictates otherwise.

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

First, a fire detection apparatus according to an embodiment of the present invention will be described with reference to FIG.

1, a fire sensing apparatus 100 according to an exemplary embodiment of the present invention includes a real camera 110, a thermal imager 120, a flame sensor 130, a data analysis processor 140, An output unit 150, and an alarm unit 160.

The real image camera 110 can be installed in an area to be monitored, and can capture real images of a monitored area to generate image data.

For example, the radiographic camera 110 may be a CCD (Charge-Coupled Device) camera, but is not limited thereto.

The radiographic camera 110 can sense fog in the area to be monitored.

In one embodiment, the real camera 110 may be configured to generate photographed image data when motion is detected in a region to be photographed, that is, a region to be monitored.

That is, the real camera 110 may be configured to store the photographed image to generate the image data when the motion due to the mist or the like is detected, without storing the moving image when there is no motion in the monitored area.

In addition, the thermal imaging camera 120 may be installed in a region to be monitored, and can capture thermal images of the region to be monitored and generate image data.

For example, the thermal imager 120 may comprise an infrared camera, but is not limited thereto.

The thermal imaging camera 120 can sense the temperature of the monitored area, i.e., heat.

In addition, the flame sensor 130 may sense the flame of the monitored area.

For example, the flame detector 130 may include an infrared flame detector IR3 that detects the infrared wavelength characteristic emitted from the flame and operates when the change of the infrared ray reaches a predetermined amount or more. However, the present invention is not limited thereto.

In one embodiment, the radiographic camera 110, the thermal imaging camera 120, and the flame detector 130 may be integrated into a single case, but the present invention is not limited thereto.

The data analysis processing unit 140 may collect and analyze the image data photographed by the radiographic camera 110 and the thermal imaging camera 120 and the flame detection signal of the flame sensor 130.

The data analysis processing unit 140 determines that a fire has occurred when at least two of the fog, high temperature, and flame are generated in the monitored area. The data analysis processing unit 140 determines the occurrence of a fire based on the output unit 150 and the alarm unit 160, Can be controlled.

As described above, the data analysis processing unit 140 determines that a fire is generated when at least two of fog, high temperature, and flame are generated. Therefore, if any one of the fog, the high temperature, and the flame is generated, the data analysis processing unit 140 may not judge the occurrence of a fire.

Accordingly, the data analysis processing unit 140 may be configured to perform the first analysis of the first fire type in which fog and high temperature are generated, the second fire type in which high temperature and spark occurs, the third fire type in which fog and spark are generated, 4 Fire type can be judged as fire.

The image analysis and the specific fire determination operation of the data analysis processing unit 140 will be described later.

Meanwhile, the output unit 150 may receive a video signal from the data analysis processing unit 140 and output the video signal to a screen of a terminal that the user can view with the naked eye.

In an embodiment, the output unit 150 may display a visual warning to the user through a pop-up window or a screen flicker when the data analysis processor 140 receives a fire occurrence command.

Also, in one embodiment, the output unit 150 may be configured to remotely transmit the processed image and data in the data analysis processing unit 140 to the portable terminal of the user using wireless communication.

If it is determined that a fire has occurred in the data analysis processing unit 140, the alarm unit 160 may receive a command from the data analysis processing unit 140 and execute a warning.

In one embodiment, the alarm unit 160 may be configured to remotely display a fire warning alert on the user's portable terminal using wireless communication, similar to the output unit 150. [

Meanwhile, although not shown, the fire detection apparatus 100 according to an embodiment of the present invention may include a database server that can store the fire occurrence status determined by the data analysis processing unit 140 as an image file and a moving picture file.

Next, a real image analysis method of the data analysis processing unit 140 will be described with reference to FIGS. 2 to 8. FIG.

2, the data analysis processing unit 140 analyzes the image data photographed by the real camera 110 to detect the magnitude of the magnitude (S208), compares the magnitude of the detected magnitude with a predetermined value, (S209).

In addition, the data analysis processing unit 140 may output the foggy image to the output unit 150 and operate the alarm unit 160 when the fog due to the fire is detected and the detection signal of the flame sensor 130 is received .

That is, in the case of the third fire type where fog and flame are generated in the monitored area, the fire detection apparatus 100 according to the embodiment of the present invention can fire alarm through the method shown in FIG.

Specifically, as shown in FIG. 2, the data analysis processing unit 140 can receive real image data from the real image camera 110 (S201).

Thereafter, the data analysis processing unit 140 may extract the boundaries of the haze by adjusting brightness and contrast of the received image (S202).

3, the data analysis processing unit 140 can obtain the image shown in FIG. 3 (b) by adjusting the brightness and contrast of the image of FIG. 3 (a) , The boundary of the object shown in FIG. 3 (c) can be extracted through the image shown in FIG. 3 (b).

In addition, the data analysis processing unit 140 may apply the same color to neighboring regions in the image to recognize fog dispersed in different colors (S203).

In addition, the data analysis processing unit 140 may remove the noise of the image to distinguish the non-fog and the non-fog as shown in the example photograph of Fig. 4 (S204). Here, FIG. 4A shows an image before noise removal, and FIG. 4B shows an image after noise removal.

For example, a Gaussian filter may be used to remove image noise, but not limited thereto, and a blur filter, a median filter, or the like may be used.

In addition, the data analysis processing unit 140 may perform shape-based processing on the image after removing the noise of the image (S205). Image feature-based processing can remove features (noise, incomplete edges, etc.) of small image objects in the image.

In one embodiment, the shape-based processing includes an erosion process (b) for removing the protruding portion of the original image (a) and an expansion process for filling the recessed portion of the original image (a) (c) can be used.

Thus, the data analysis processing unit 140 can generate the shape-based processed image of FIG. 6 (b) from the original image of FIG. 6 (a).

Meanwhile, the data analysis processing unit 140 may remove noise by comparing the image with a predetermined threshold value after the shape-based processing of the image (S206).

That is, the data analysis processing unit 140 compares the brightness of the edge of the object with the preset brightness threshold in the original image of FIG. 7 (a), removes the darker portion than the threshold value, Processing can be performed to generate the noise-free image of Fig. 7 (b).

In addition, the data analysis processing unit 140 can perform a process of detecting a boundary of a target object after the image has been removed (S207).

That is, the data analysis processing unit 140 can generate an image in which the boundary of the object is expressed as shown in FIG. 8 (b) by detecting the boundary of the object in the noise-removed original image of FIG. 8 (a).

Thereafter, the data analysis processing unit 140 can detect the magnitude of the fog through the image shown in FIG. 8B (S208). The magnitude of the detected fog is compared with the preset data, (S209).

If the magnitude of the detected fog is smaller than the preset data, it is not determined that the fog is caused by fire, and the process can return to the step S201 for receiving the real image again.

Alternatively, when the magnitude of the detected fog is larger than the preset data, the data analysis processing unit 140 may determine that the fog is caused by a fire and track the shape and direction of the fogging (S210 ).

When the position of the flame matches the position of the flame, the data analysis processing unit 140 determines that the flame is in a fire state and outputs the result to the output unit 150 and the alarm unit 160. [ (S230).

At this time, the data analysis processing unit 140 determines that the fog and the flame are not matched, and judges that the fog is not fired. Instead, the data analysis processing unit 140 simply outputs the warning to the output unit 150 without issuing a warning display command to the output unit 150 and the alarm unit 160 It is possible to output a screen on which haze is photographed.

Next, with reference to FIG. 9, a method of analyzing a real image and a thermal image of the data analysis processing unit 140 will be described.

9, the data analysis processing unit 140 analyzes a real image photographed by the real image camera 110 and a thermal image photographed by the thermal image camera 120, The matching area of the image is extracted (S309), and output to the output unit 150 as PIP (Picture In Picture) (S312).

Here, the data analysis processing unit 140 outputs a matching area as an image to the output unit 150 when it is determined that the matching area (the part where the object of the real image matches the object of the thermal image) is extracted , The alarm unit 160 can be operated.

That is, in the case of the first fire type in which fog and high temperature occur in the monitored area, the fire detection apparatus 100 according to an embodiment of the present invention can fire alarm through the method shown in FIG.

As shown in FIG. 9, the data analysis processing unit 140 performs a data analysis process based on the real image captured by the real camera 110 and the thermal image captured by the thermal camera 120 (S301, S302).

Thereafter, the data analysis processing unit 140 may apply a color map to the real image and the thermal image (S303).

Here, the color map means a process of converting an infrared image, which is a grayscale image, into a color image.

After applying the color map to the real image and the thermal image, the data analysis processing unit 140 adjusts brightness and contrast (S304), noise removal (S305), shape based processing (S306) Threshold value comparison noise removal (S307) and boundary detection processing (S308) can be performed.

Here, the brightness contrast adjustment (S304), noise reduction (S305), shape based processing (S306), threshold comparison noise elimination (S307), and boundary detection processing (S308) They are substantially the same, and a description thereof will be omitted.

On the other hand, the data analysis processing unit 140 can extract the matching area between the real image and the thermal image by comparing the image that detects the boundary between the real image and the object of the thermal image through the image processing (S309 ).

Thereafter, the data analysis processing unit 140 can determine whether the real image and the thermal image are matched (S310). If the real image and the thermal image do not match, the data analysis processing unit 140 determines that the real image and the thermal image are non- And returning to steps S301 and S302 for receiving the thermal image.

On the contrary, when the real image and the thermal image are matched, the data analysis processor 140 can analyze the length and the ratio of the matching area (S311).

Thereafter, the data processing unit can instruct the output unit 150 to process the PIP (Picture In Picture) so that the user can compare the real image and the thermal image and easily know whether the image is a fire or not.

At this time, data analyzing the length and the ratio of the matching area are displayed on the output unit 150, so that the user can know the scale of the fire and the location of the fire.

Finally, referring to FIG. 10, a method of analyzing a thermal image of the data analysis processing unit 140 will be described.

10, the data analysis processing unit 140 analyzes the thermal image captured by the thermal imaging camera 120 to detect the size of the detection object (S407), and determines the size of the detected detection target The minimum setting size is compared (S408), the holding time of the detection target is compared with a preset time (S409), and it is possible to determine whether the fire is caused by a high temperature.

Here, the data analysis processing unit 140 outputs a thermal image to the output unit 150 and operates the alarm unit 160 when a high temperature due to fire is detected and a detection signal of the flame sensor 130 is received have.

That is, in the case of the second fire type in which a high temperature and a flame are generated in the monitored area, the fire detection apparatus 100 according to an embodiment of the present invention can fire a fire alarm through the method shown in FIG.

Specifically, as shown in FIG. 10, the data analysis processing unit 140 can receive a thermal image photographed by the thermal imaging camera 120 (S401).

Thereafter, the data analysis processing unit 140 may apply the color map to the thermal image (S402).

Then, the data analysis processing unit 140 can extract the maximum temperature and the minimum temperature from the photographed thermal image (S403).

Thereafter, the data analysis processing unit 140 may set an area having a high temperature in the image as the detection area, analyze the image of the detection area, and exclude the part other than the detection area in the image analysis (S404).

By setting the detection region and analyzing the image, the range for tracking the temperature can be minimized and the image analysis efficiency can be improved.

After setting the detection area, the temperature of the detection range can be set (S405). In one embodiment, as a method of setting the temperature of the detection range, a method of displaying a portion where the temperature of the detection range in the thermal image is larger than the reference value in red may be used.

Here, the reference value may be preset by the user.

Thereafter, the data analysis processing unit 140 may set the temperature of the image to binarization (S406). In the temperature binarization method, a portion in which the temperature is lower than the reference temperature is expressed in black, and a portion in which the temperature exceeds the reference temperature is expressed in red.

Then, the data analysis processing unit 140 can detect the size of the detection object through the binarized image (S407).

Thereafter, the data analysis processing unit 140 compares the size of the detected detection target with a predetermined minimum setting size (S408). If the size of the detection target is larger than the minimum setting size, the data analysis processing unit 140 sets the time It can be compared with the predetermined time (S409).

In this case, if the object to be detected is maintained at the predetermined time or longer, the data analysis processing unit 140 may determine that the object is a fire due to a fire, and may track and track the direction and direction of the object.

On the other hand, if the size of the detection object is less than the minimum setting size or if the detection object is kept shorter than the predetermined time and disappeared, the process may return to the step S401 of receiving the thermal image.

On the other hand, after tracking the detection object, the data analysis processing unit 140 matches the detection signal of the flame sensor with the region where the high temperature is generated in the image (S411). If the position of the high temperature matches the position of the mist, The alarm output unit 150 and the alarm unit 160 may display a warning (S412).

The fire sensing apparatus 100 according to an embodiment of the present invention is capable of accurately detecting fires by collectively determining fog, heat, and flame, and does not generate fire alarm.

While the present invention has been particularly shown and described with reference to particular embodiments thereof, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art without departing from the spirit and scope of the invention as defined by the following claims I would like to make it clear.

100: Fire detector
110: radiographic camera 120: infrared camera
130: Flame detector 140: Data analysis processor
150: output unit 160: alarm unit

Claims (12)

A real image camera that captures real images to generate image data, and detects fogging;
An infrared camera for capturing a thermal image to generate image data and sensing temperature;
A flame detector for detecting flame;
The image data photographed by the real image camera and the thermal image camera and the detection signal of the flame sensor are collected and analyzed and data which is judged as a fire occurrence when at least two of mist, An analysis processing unit;
An output unit receiving the video signal from the data analysis processing unit and outputting the video signal to the screen; And
An alarm unit for executing a warning when the data analysis processing unit determines that a fire has occurred;
And a fire detector.
The method according to claim 1,
Wherein the real image camera is configured to generate photographed image data when motion is detected in an area to be photographed.
The method according to claim 1,
Wherein the data analysis processing unit analyzes the image data photographed by the real camera and detects the size of the mist and compares the detected size of the mist with a preset value to determine whether or not the image is fogged by fire.
The method of claim 3,
Wherein the data analysis processing unit outputs a smoke image to the output unit and activates the alarm unit when the smoke is detected by the fire and the detection signal of the flame sensor is received.
The method of claim 3,
Wherein the data analysis processing unit detects brightness magnitude by performing brightness contrast adjustment, noise reduction, shape-based processing, threshold value comparison noise removal, and boundary detection processing on the image data photographed by the real image camera.
6. The method of claim 5,
Wherein the data analysis processing unit erodes and expands the image in the shape-based processing.
The method according to claim 1,
The data analysis processing section analyzes the image photographed by the real image camera and the image photographed by the thermal image camera to extract a matching area between the real image and the thermal image, and outputs the matching area to a PIP (Picture In Picture) Wherein the fire detection device comprises:
8. The method of claim 7,
Wherein the data analysis processing unit outputs the matching area as an image to the output unit and activates the alarm unit when the matching area is extracted.
8. The method of claim 7,
The data analysis processing section performs brightness contrast adjustment, noise removal, shape-based processing, threshold value comparison noise removal, and boundary detection processing on the image photographed by the real image camera and the image photographed by the thermal image camera, And extracts a matching area of the image image.
The method according to claim 1,
The data analysis processing unit analyzes the image captured by the thermal imaging camera to detect the size of the detection object, compares the detected size of the detection object with the minimum setting size, compares the detection time of the detection object with a predetermined time, Is judged to be a high temperature caused by the fire.
11. The method of claim 10,
Wherein the data analysis processor outputs a thermal image to the output unit and operates the alarm unit when a high temperature due to a fire is detected and a detection signal of the flame sensor is received.
11. The method of claim 10,
Wherein the data analysis processing unit detects a size of a detection target by applying a color map, extracting a maximum temperature and a minimum temperature, setting a detection region, setting a detection range temperature, and setting a temperature binarization on an image taken by the thermal imaging camera Fire detection device.

Images