JP6139977B2 - Fire detection device and fire detection method - Google Patents

Description

The present invention relates to a fire detection device and a fire detection method for detecting smoke in an early stage of fire from an image of a monitoring area captured by a monitoring camera.

  Conventionally, various devices and systems have been proposed in which a fire is detected by performing image processing on an image of a monitoring area captured by a monitoring camera.

  In such a fire detection device, early detection of a fire is important from the viewpoint of initial extinguishing and evacuation guidance for the occurrence of a fire.

For this reason, in the conventional device (Patent Document 1), as a phenomenon caused by smoke from a fire from an image, a decrease in transmittance or contrast, a convergence of a luminance value to a specific value, a luminance distribution range is narrowed and a luminance distribution is reduced. , A change in average value of brightness due to smoke, a decrease in the total amount of edges, and an increase in intensity in the low frequency band are derived, and these can be comprehensively judged to detect smoke.

JP 2008-046916 A JP-A-7-245757 JP 2010-238028 A

  By the way, in the smoldering combustion which is often in the early stage of fire, a very small amount of smoke rises, the amount of smoke increases with the passage of time, and finally a smoke layer is generated along the ceiling surface. Detects smoke layers generated on the ceiling.

  In this way, it is important that a very small amount of smoke rises and detect the fire at the early stage of the fire. However, in a conventional device for detecting smoke by performing image processing on an image, for example, rising smoke The movement (flow) is detected, but it is difficult to detect unless a sufficient amount of smoke rises to detect this flow, and it becomes a very small amount like a thin line. There was a problem that it could not be detected in the early stage of the fire where the smoke of the fire rose.

It is an object of the present invention to provide a fire detection device and a fire detection method that can detect smoke in a smoldering fire and an initial fire by image processing.

(apparatus)
The present invention provides a fire detection device,
An imaging means for capturing an image of the monitoring area;
Ridge line extraction means for extracting a ridge line from an image captured by the imaging means and generating a ridge line image;
A difference processing means for generating a ridge line difference image from a ridge line image at a predetermined time interval generated by the ridge line extraction means;
A fire determination means for detecting a ridge line formed by smoke from the ridge line difference image generated by the difference processing means and determining a fire;
It is provided with.

In another embodiment of the present invention, in the fire detection device,
An imaging means for capturing an image of the monitoring area;
Difference processing means for generating a difference image from images of a predetermined time interval imaged by the imaging means;
A ridge line extracting means for extracting a ridge line from the difference image generated by the difference processing means and generating a ridge line difference image;
A fire judging means for judging a fire by detecting a ridge line formed by smoke from the ridge line difference image generated by the ridge line extracting means;
It is provided with.

Here, the ridge line extraction means performs edge enhancement processing on the image to extract the ridge line,
Fire judgment means
A linear component extraction unit that divides the ridge line difference image into a plurality of image regions and extracts a linear component of the ridge line for each image region;
For each image region, time-series change detecting means for detecting a time-series change of the linear component extracted by the linear component extracting means,
A fire detection means for determining a fire when a characteristic predetermined time series change due to smoke is detected from the time series change of the linear component detected by the time series change detection means;
Is provided.

The linear component extraction means extracts the inclination and occurrence frequency of the linear component for each image area,
Time series change detecting means detects a time-series change in the tilt and frequency of the linear component extracted by the linear component extracting means,
The fire detection means, as characteristic time-series changes due to smoke, is characterized by the slope and occurrence frequency of the characteristic linear component due to smoke from the time-series change of the slope and frequency of occurrence of the linear component detected by the time-series change detection means. A fire is determined when a predetermined time-series change is detected.

  The fire detection means determines that a fire has occurred when one or more time-series changes of linear components having a constant inclination and different occurrence frequencies are detected.

(Method)
The present invention provides a fire detection method,
An image of the monitoring area is captured by the imaging means,
A ridge line is extracted from the image captured by the imaging unit by the ridge line extraction unit to generate a ridge line image,
A difference processing means generates a ridge line difference image from a ridge line image of a predetermined time interval generated by the ridge line extraction means,
From the ridge line difference image generated by the difference processing means, the fire determination means detects the ridge line formed by smoke, and determines the fire.
It is characterized by that.

In another embodiment of the present invention, in the fire detection method,
An image of the monitoring area is captured by the imaging means,
A difference processing unit generates a difference image from a predetermined time interval image captured by the imaging unit,
Extract the ridge line from the difference image generated by the difference processing means by the ridge line extraction means to generate a ridge line difference image,
From the ridge line difference image generated by the difference processing means, the fire determination means detects the ridge line formed by smoke, and determines the fire.
It is characterized by that.

The other features of the fire detection method of the present invention are basically the same as those of the above-described fire detection device, and thus description thereof is omitted.

  According to the fire detection device and the fire detection method of the present invention, a ridge line image is generated by extracting a ridge line generated by smoke from an image of a monitoring area captured by an imaging unit, and a ridge line image at a predetermined time interval. As a result of generating a ridge line difference image by extracting the difference of the ridge line, extracting a linear component of the ridge line from this ridge line difference image, and obtaining a change in time series with respect to the inclination and occurrence frequency, a characteristic change due to smoke is obtained. By detecting this, it is possible to reliably detect the start of a very small amount of smoke generated by a burning fire or a cigarette discarded in a trash bin, and to make an early judgment and notify the fire. .

  In addition, when extracting the ridge line from the image, the ridge line by the background other than smoke is also extracted, but by taking the difference of the ridge line image at a predetermined time interval, the ridge line extracted from the background is deleted and the ridge line by smoke Since only smoke is extracted and the smoke extends upward while fluctuation, the characteristic change of the linear component due to smoke is detected by obtaining the time-series change in the slope and frequency of the linear component extracted from the ridgeline. Allows judgment of fire.

In addition, when taking a difference between ridgeline images at predetermined time intervals, the ridgelines extracted from smoke that fluctuates upward while moving are different in position in time, so the difference image includes the previous image that has disappeared at the present time. Both the smoke ridge line of the current image and the smoke ridge line of the current image that has newly appeared (two smoke ridge lines) are generated, and the straight line component of the ridge line that is approximately twice that of the case where the difference is not taken is extracted. It is possible to improve the certainty of fire judgment by emphasizing the change of simple linear components

Explanatory drawing which showed the monitoring area | region where the fire detection apparatus of this invention was installed An explanatory diagram modeling the state where a very small amount of smoke rises Block diagram showing outline of functional configuration of image processing apparatus Explanatory drawing showing erasing of background ridge line by difference processing Explanatory diagram showing image segmentation Explanatory diagram showing an example of a ridgeline image that changes in time series by cutting out the fire source part in a state where a very small amount of smoke rises, and a ridgeline differential image generated by the difference processing Explanatory drawing showing the temporal change of the ridge line difference image with a very small amount of smoke rising, along with the area directly above the fire source Explanatory drawing showing the time-series change of the linear component of the area directly above the fire source in the ridge line difference image Explanatory drawing which showed the time series change of the vector showing the inclination of the linear component of FIG. 8, and the occurrence frequency Explanatory drawing which showed the time-sequential change of the linear component of the area | region above the area | region of FIG. Explanatory drawing which showed the time-sequential change of the vector showing the inclination of linear component of FIG. 10, and the occurrence frequency. Flowchart showing the operation of the image processing apparatus of FIG. The block diagram which showed the outline of the function structure of the image processing apparatus used as other embodiment

[Outline of fire detection device]
FIG. 1 is an explanatory view showing a monitoring area where a fire detection device according to the present invention is installed. FIG. 1 (A) shows a side view, and FIG. 1 (B) shows a front view seen from a monitoring camera.

  As shown in FIG. 1A, a monitoring camera 10 functioning as an imaging unit is installed in the monitoring area 16, and an image is obtained by imaging the state of the monitoring area shown in FIG.

  A combustible material placed in the monitoring area 16 causes a fire for some reason, and a very small amount of smoke 24 rises as a thin line at the beginning of the fire as shown in the figure. Moreover, the wall surface of the monitoring region 16 appears as a straight line in the vertical direction or the horizontal direction due to the structure or wallpaper.

An image captured by the monitoring camera 10 is transmitted to an image processing apparatus 12 installed in a manager's room or the like via a transmission path, and a small amount of smoke 24 rising from a fire source 18 such as a trash can is detected by image processing. A fire is judged, and a fire detection signal is output to the fire alarm facility 14 to output a fire alarm.

[Detection principle]
The principle of detecting a small amount of smoke according to the present invention will be described as follows. In the present invention, a small amount of smoke 24 rising from the fire source 18 shown in FIG. 2A is detected by image processing. In this case, the initial smoke 24 is translucent, as shown in FIG. It can be considered as a smoke model 25 in which a cylindrical object extends upward from the fire source 18 while fluctuating.

  In the smoke model 25, as shown in the density distribution of FIG. 2C, the smoke density is higher at the center and relatively thinner at the periphery. It is thought that the center of the line draws the ridgeline that is least transparent.

  Thus, edge enhancement processing is applied to the captured image to extract a smoke ridgeline, and then a ridgeline differential image is generated from two ridgeline images at a predetermined time interval. By generating this ridge line difference image, the ridge line of the background that exists constantly (fixed) is deleted, and only the smoke ridge line remains. Subsequently, after the ridge line difference image is divided into fine regions, the Hough transform is performed on each region to extract the linear component of the smoke ridge line. The linear component due to the smoke extracted in this way extends upward while fluctuating over time. For this reason, if the change of the extracted linear component in the time series is caught, the characteristic time series change due to the smoke can be caught.

Based on the above results, the time-series changes in the direction and frequency of the linear component of each region in the difference image of the ridge line extracted from the image captured at every predetermined period, characteristic time-series changes due to smoke Thus, it is possible to detect a small amount of smoke 24 that rises as a thin line in the early stage of a fire.

[Fire detection device]
(Functional configuration of fire detection device)
FIG. 3 is a block diagram schematically showing the functional configuration of the fire detection device according to the present invention. As shown in FIG. 3, the fire detection apparatus includes a monitoring camera 10 and an image processing apparatus 12. The image processing apparatus 12 includes a computer circuit having a CPU, a memory, various input / output ports and the like as hardware. As functions realized by execution of a program by the CPU, a ridge line extraction unit 28 that functions as a ridge line extraction unit, a difference processing unit 30 that functions as a difference processing unit, and a fire determination unit 32 that functions as a fire determination unit are provided. Furthermore, as a function of the fire determination unit 32, a linear component extraction unit 34 that functions as a linear component extraction unit, a time series change detection unit 36 that functions as a time series change detection unit, and a fire detection unit 38 that functions as a fire detection unit. Is provided. The transmission unit 26 uses an appropriate transmission interface that receives image data captured by the monitoring camera 10.

  The monitoring camera 10 functioning as an imaging unit transmits image data of a monitoring area, for example, 30 frames per second as moving image data by transmission control of the transmission unit 26 and stores it in a memory (not shown) provided in the image processing device 12. .

  The ridge line extraction unit 28 generates a ridge line image by extracting a ridge line from the frame-unit image stored in the memory. For example, in this case, the ridge line extraction unit 28 applies a Sobel filter that is one of the edge enhancement processes to the image, and extracts a ridge line, for example, as shown in the ridge line image 20 of FIG. To generate a ridge line image. Note that the ridge line extraction processing by the ridge line extraction unit 28 may be performed for each frame in which a predetermined number of frames are thinned out because of the processing speed, without targeting all frame images.

  The Sobel filter is present in the image by multiplying the nine pixel values at the upper left and right with a certain target pixel as the center by multiplying a predetermined coefficient by two coefficient matrices in the horizontal and vertical directions to obtain the sum. This is a differential process that enables detection of the boundary (edge) of a certain region. By applying this, the ridge line extraction unit 28 detects smoke 24 from the smoke image rising upward from the fire source 18 shown in FIG. Extract edge of.

  The difference processing unit 30 generates a ridge line difference image from two ridge line images at a predetermined time interval generated by the ridge line extraction unit 28. FIG. 4 is an explanatory diagram showing the difference processing when smoke does not exist, and the image generated in chronological order from the ridge line extraction unit 28 is extracted from the background ridge line as shown in FIG. 4A. A ridge line image 20 is obtained. Since the difference processing unit 30 takes the difference between two adjacent ridge line images 20 that are not changed in FIG. 4A, the ridge line difference image in which the background ridge line is deleted as shown by the dotted line in FIG. 4B. 21 is generated.

  FIG. 5 shows the ridge line images F1 to F5 generated for each predetermined thinning frame period by the ridge line extraction unit 28 in the situation where the small amount of smoke 24 shown in FIG. It is explanatory drawing which took out the one part screen area | region which extracted the ridgeline difference image F12-F45 produced | generated by the part 30 including a fire source.

  First, the ridge line image F1 is an image immediately before the smoke rises, and the next ridge line image F2 is an image in which the smoke starts to rise first. The ridge line difference image F12 calculates the difference from the previously generated ridge line image F1 in units of pixels at the timing when the ridge line image F2 is generated, and sets a pixel value equal to or greater than a predetermined threshold to an effective difference pixel (for example, a black pixel value). ) To generate the ridge line difference image F12. In this ridge line difference image F12, the fixed background ridge line is erased like a dotted line, and only the newly appearing smoke ridge line 24a is extracted.

  Subsequently, when the ridge line image F3 is generated, a difference process with the previously generated ridge line image F2 is performed to generate a ridge line difference image F23. In this ridge line difference image F23, the fixed background ridge line is erased like a dotted line, the newly appearing smoke ridge line 24b is extracted, and the smoke ridge line 24a that appears last time is extracted. That is, in the ridge line difference image F23, both the smoke ridge line 24a that appeared last time but has disappeared now and the smoke ridge line 24b that has newly appeared are generated. In addition, the overlapping part of the smoke ridge lines 24a and 24b is deleted.

  Hereinafter, even when the ridge line images F4 and F5 are generated, the same difference process generates the ridge line difference images F34 and F45 including the smoke ridge line that appeared last time but is now disappeared and the newly appeared smoke ridge line. Is done.

  The fire determination unit 32 in FIG. 3 determines a fire by detecting a specific ridge line characteristic of smoke from the ridge line difference image extracted by the difference processing unit 30. Specifically, the fire component extraction is performed. The fire is detected by the unit 34, the time-series change detection unit 36, and the fire detection unit 38.

  The straight line component extraction unit 34 divides the ridge line difference image generated by the difference processing unit 30 into a plurality of regions, for example, 64 × 64 pixel regions, as shown by broken lines in FIG. Hough transform) is performed to extract the straight line component of the ridge line 24a. The Hough transform is known as a method for extracting a straight line segment in an image. For n points in the image, n curves are obtained on the ρ-θ plane. Of these, m curves are obtained. Are intersected at one point, m points on the image corresponding to the m points are on the same straight line, and thereby a linear component can be extracted.

  The time series change detection unit 36 obtains the time series change of the slope and the occurrence frequency of the linear component of the smoke ridge line 24a extracted by the Hough transform by the linear component extraction unit 34.

  FIG. 7 is an explanatory diagram showing the whole of the ridge line difference images F12 and F23 of FIG. 5. As an area to be processed, an area A1 immediately above the fire source 18 and an area A2 above it are shown as an example. .

  In the difference area image F12 in FIG. 7A, the smoke ridge line 24a rises from the fire source 18 and the tip is in the area A1, and in the next difference area image F23 in FIG. Further rising and passing through the areas A1 and A2, the previous smoke ridge line 24a is generated. On the other hand, the background ridge line that does not change with time disappears as shown by the dotted line.

  FIG. 8 shows a time-series change of the linear component extracted from the smoke ridge line of the ridge line difference image at times t1 to t4 for each predetermined thinning frame period in the area A1 immediately above the fire source 18 in FIG. At time t1, the linear component of the smoke ridge line passing through the area A1 has two occurrence frequencies when the upper direction is θ1 = 0 ° in the two-dimensional coordinates with the center of the lower side of the area as the origin, and the upper right diagonal direction is θ2. The occurrence frequency is two, and if the upper left diagonal is θ3, the occurrence frequency is one.

  The direction and frequency of occurrence of the linear component of the smoke ridge line passing through the region A1 changes as shown at times t2 to t4 according to the fluctuation of the rising smoke. In addition, at time t2 to t4, both the straight line component of the previous smoke ridge line and the current smoke line component exist, and therefore, the line component of two smoke ridge lines is extracted for the case of the ridge line image that is not subjected to differential processing. By doing so, the frequency of occurrence increases approximately twice, and the characteristic change of the linear component due to smoke can be emphasized, which contributes to improving the certainty of fire judgment.

  FIG. 9 is an explanatory diagram showing a time-series change of the linear component in the area A1 of FIG. 8. The linear component is shown by accumulating vectors having a slope θ and the length of occurrence frequency.

As shown in FIG. 9, at time t1, the vector B1 is (θ1, 2).
The vector B2 is (θ2, 2)
The vector B2 is (θ3, 1)
Thus, from time t2 to t4, it increases cumulatively according to the time series change.

  FIG. 10 shows a time-series change of the linear components extracted at times t1 to t4 for each predetermined thinning frame period in the area A2 above the area A1 immediately above the fire source 18 in FIG. In the area A2, the rising smoke fluctuates, and therefore, at time t1, the linear component of the smoke ridge line passing through the area A2 has an occurrence frequency of 1 when the upper direction is θ1 = 0 °, and the upper right diagonal If θ2 is θ2, the occurrence frequency is 1, and if the upper left is θ3, the occurrence frequency is 2. Further, if the upper right oblique direction is larger than θ2, the occurrence frequency is one, and the upper left oblique direction is larger than θ3. If θ5, the frequency of occurrence is one.

  Thus, the direction and frequency of occurrence of the linear component of the smoke ridge line passing through the region A2 changes as shown at times t2 to t4 according to the fluctuation of the rising smoke. In addition, at times T2 to t4, both the previous smoke ridge line linear component and the current smoke linear component exist.

  FIG. 11 is an explanatory diagram showing a time-series change of the linear component in the area A2 of FIG. 10, and shows the linear component accumulated by a vector having a slope θ and a length of occurrence frequency.

As shown in FIG. 11, at time t1, the vector B1 is (θ1, 1),
The vector B2 is (θ2,1)
The vector B3 is (θ3, 2),
The vector B4 is (θ4, 1),
The vector B5 is (θ5, 1)
Thus, from time t2 to t4, it increases cumulatively according to the time series change.

  The fire detection unit 38 in FIG. 3 detects a characteristic time-series change due to smoke from the time-series changes of the linear components of each region detected by the time-series change detection unit 36, and determines a fire.

  For example, when the time-series change detection unit 36 detects a time-series change given by the accumulation of the slope of the linear component and the occurrence frequency shown at time t4 in FIGS. 9 and 11, the characteristic time-series change due to smoke is 7 and 9, a so-called daisy pattern in which a plurality of cumulative vectors different in slope and frequency of occurrence exists in a so-called daisy pattern, and the fire detection unit 38 has a characteristic time-series change due to smoke. A fire is detected by detecting a daisy pattern indicating.

Further, when a person moves in the monitoring area, for example, a divided area having a daisy pattern as shown at time t1 in FIG. 11 is temporarily generated, but a cumulative vector that is a time-series occurrence frequency does not increase, By setting a predetermined threshold value for the occurrence frequency and setting the occurrence frequency equal to or higher than the threshold value as the object of smoke determination, the linear component of the ridge line accompanying the movement of the person can be eliminated.

  In addition, when a person remains in the monitoring area, it can be erased as a background ridge line by the ridge line image difference process, which does not affect the fire determination.

[Fire judgment operation]
FIG. 12 is a flowchart showing a fire detection operation by the image processing apparatus of FIG.

  In FIG. 12, the image processing device 12 acquires an image of a monitoring area captured at, for example, 30 frames / second as a moving image by the monitoring camera 10 in step S1 (hereinafter, “step” is omitted), and stores it in the memory. Then, by applying the Sobel filter by the ridge line detection unit 28, a ridge line is extracted from an image extracted at a predetermined thinning frame period to generate a ridge line image.

  Subsequently, in S3, the difference processing unit 30 subtracts the previously generated ridge line image from the ridge line image generated this time to generate a ridge line difference image.

Subsequently, in S4, the straight line component extraction unit 34 divides the ridge line difference image into a plurality of regions, and in S5, Hough transform is performed for each region to extract the straight line component of the ridge line, and then the process proceeds to S6 and the time series change detection unit 36 To obtain the time-series change due to the slope and occurrence frequency of the linear component extracted in S5, and in S7, the fire detection unit 36 uses the slope of the linear component and the slope of the characteristic linear component due to smoke from the time-series change of the occurrence frequency. When a fire is detected to detect a predetermined time-series change that occurs , for example, a daisy pattern, and as a result, a fire is detected at S8, a fire detection signal is output to the fire alarm facility at S9, and a fire alarm is output. Let On the other hand, if no fire is detected in S9, the process returns to S1 and the same processing is repeated.

[Other Embodiments of Image Processing Apparatus]
FIG. 13 is a block diagram showing an outline of a functional configuration of a fire detection device according to another embodiment. As shown in FIG. 13, the fire detection device includes a monitoring camera 10 and an image processing device 12. The image processing device 12 includes a difference processing unit 30, a ridge line extraction unit 28, and a fire determination unit 32 that functions as a fire determination unit. Furthermore, as a function of the fire determination unit 32, a linear component extraction unit 34, a time series change detection unit 36, and a fire detection unit 38 are provided. The transmission unit 26 uses an appropriate transmission interface that receives image data captured by the monitoring camera 10.

  This embodiment is characterized in that a difference processing unit 30 is provided upstream of the ridge line extraction unit 28 with respect to the embodiment of FIG. Therefore, the difference processing unit 30 generates a difference image from adjacent images stored in time series in the memory. By generating a differential image by this differential processing, it is possible to erase a background image that exists constantly and extract an image of a very small amount of smoke that rises as a fine streak at the beginning of a fire.

  Further, since spike-like noise may be extracted from the difference image generated by the difference processing unit 30, processing for removing isolated points is performed on this.

  The ridge line extraction unit 28 is different in that a ridge line difference image is generated by extracting a ridge line from the difference image generated by the difference processing unit 30, but the ridge line extraction process is basically the same as the embodiment of FIG.

  Moreover, since the linear component extraction part 34 provided in the fire judgment part 32, the time series change detection part 36, and the fire detection part 38 become the same as embodiment of FIG. 3, the description is abbreviate | omitted.

  Since the ridge line extraction unit 28 performs the ridge line extraction process on the difference image generated by the difference processing unit 30, when applying a Sobel filter, which is one of the edge enhancement processes, the difference image occupying the entire screen is For example, for invalid pixels (erased pixels) that have only a small screen portion such as a smoke image and do not have a pixel value generated by difference processing, by applying no Sobel filter as the target pixel, The burden can be greatly reduced, and ridge lines can be extracted at high speed.

[Modification of the present invention]
(Ridge line extraction)
In the above embodiment, the sobel filter is applied to the image to extract the edge of the smoke, but an appropriate filter used for edge enhancement such as a Previtt filter is applied. Also good.

(Linear component extraction)
In the above embodiment, the smoke ridge line is extracted by applying the Hough transform. However, a processing method for extracting a linear component from an image such as Line Segment Detector (LSD) may be applied.

(Image processing device)
In the above embodiment, the surveillance camera and the image processing apparatus are separately arranged and connected by a transmission path, but an apparatus in which both are integrated may be used.

The present invention is not limited to the above-described embodiment, includes appropriate modifications without impairing the object and advantages thereof, and is not limited by the numerical values shown in the above-described embodiment.

10: surveillance camera 12: image processing device 14: fire alarm equipment 16: monitoring area 18: fire source 20: ridge line image 24: smoke 25: smoke model 26: transmission unit 28: ridge line extraction unit 30: difference processing unit 32: fire Determination unit 34: linear component extraction unit 36: time-series change detection unit 38: fire detection unit

Claims (10)

An imaging means for capturing an image of the monitoring area;
A ridge line extracting unit that extracts a ridge line from an image captured by the imaging unit and generates a ridge line image;
A difference processing means for generating a ridge line difference image from a ridge line image of a predetermined time interval generated by the ridge line extraction means;
A fire judgment means for judging a fire by detecting a ridge line formed by smoke from the ridge line difference image generated by the difference processing means;
Fire detection apparatus characterized by comprising a.
An imaging means for capturing an image of the monitoring area;
Difference processing means for generating a difference image from images of a predetermined time interval imaged by the imaging means;
A ridge line extracting means for extracting a ridge line from the difference image generated by the difference processing means and generating a ridge line difference image;
A fire judgment means for judging a fire by detecting a ridge line formed by smoke from the ridge line difference image generated by the ridge line extraction means;
Fire detection apparatus characterized by comprising a.
In the fire detection device according to claim 1 or 2,
The ridge line extraction means performs edge enhancement processing on the image to extract a ridge line,
The fire determination means is
Dividing the ridge differential image into a plurality of image regions, for each of the image areas, and the straight line component extracting means for extracting a linear component of said ridge,
Time-series change detecting means for detecting a time-series change of the linear component extracted by the linear component extracting means for each image region;
A fire detection means for determining a fire when detecting a characteristic time-series change due to smoke from the time-series change of the linear component detected by the time-series change detection means;
A fire detection device comprising:
In the fire detection device according to claim 3,
The line component extracting means, for each of the image area, extracts the slope and frequency of the line component,
The time-series change detecting means detects a time-series change in the tilt and frequency of the linear component extracted by the linear component extracting means,
The fire detection means detects a time-series change in a given that the slope and frequency characteristic linear component due to the smoke from the time-series change in the tilt and frequency of line component detected by the time-series change detecting means If it is judged to be a fire,
A fire detection device characterized by that.
5. The fire detection device according to claim 4, wherein the fire detection means determines that a fire has occurred when one or more time-series changes of linear components having a constant inclination and different occurrence frequencies are detected. Fire detection device.
An image of the monitoring area is captured by the imaging means,
A ridge line is extracted from the image captured by the imaging unit by a ridge line extraction unit to generate a ridge line image;
A ridge line difference image is generated by a difference processing means from a ridge line image of a predetermined time interval generated by the ridge line extraction means,
From the ridge line difference image generated by the difference processing means, a fire determination means detects a ridge line formed by smoke, and determines a fire.
A fire detection method characterized by that.
An image of the monitoring area is captured by the imaging means,
A difference processing unit generates a difference image from a predetermined time interval image captured by the imaging unit,
A ridge line is extracted from the difference image generated by the difference processing means by the ridge line extraction means to generate a ridge line difference image,
From the ridge line difference image generated by the ridge line extraction means , a fire determination means detects a ridge line formed by smoke, and determines a fire.
A fire detection method characterized by that.
In the fire detection method according to claim 6 or 7,
The image picked up by the image pickup means is subjected to edge enhancement processing by the ridge line extraction means to extract a ridge line,
The fire determination means is
Said ridge difference image is divided into a plurality of image regions, for each of the image area, it extracts the linear component of said ridge,
For each of the image areas, it obtains a time-series change of the line component and the extracted,
A fire detection method characterized in that a fire is determined when a characteristic time-series change due to smoke is detected from the time-series changes of the linear component .
In the fire detection method according to claim 8,
The fire determination means is
For each of the image area, it extracts the slope and frequency of the line component,
Obtains time-series change in the tilt and frequency of the linear component and the extracted,
It is determined that a fire occurs when a predetermined time-series change that is a slope and occurrence frequency of a characteristic linear component due to smoke is detected from the time-series change of the slope and occurrence frequency of the linear component,
A fire detection method characterized by that.
  10. The fire detection method according to claim 9, wherein the fire determination means determines that a fire has occurred when one or more time-series changes of linear components having a constant slope and different occurrence frequencies are detected. Fire detection method.

Images