JP2010097265A - Smoke detecting apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a smoke detecting apparatus suppressing erroneous detection due to disturbance without limiting a monitorable area. <P>SOLUTION: The smoke detecting apparatus includes: a smoke feature quantity calculation means 30 extracting a feature quantity relating to smoke for each of a plurality of areas set inside an image captured by a monitoring camera 1; a storage part 20 storing predetermined reference determination values for determining presence/absence of smoke occurrence as a plurality of values having a plurality of detection sensitivities for each of the respective areas; an area sensitivity setting means 40 for setting a desired detection sensitivity for each of a plurality of areas according to a monitoring target by the monitoring camera 1; and a smoke determination means 50 selecting a predetermined reference determination value corresponding to a desired detection sensitivity from a plurality of predetermined reference determination values, comparing the selected value with the feature quantity, and detecting occurrence of smoke with the desired detection sensitivity in each of a plurality of areas based on the comparison result. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a smoke detection device that detects the generation of smoke by performing image processing on an image captured by a surveillance camera, and in particular, smoke that can prevent false detection and non-detection due to the influence of a disturbance. The present invention relates to a detection device.

  From the viewpoint of initial fire extinguishment in the event of a fire or prevention of escape delay in a fire accident, early detection of fire or smoke is very important. Thus, in the field of smoke detection devices, research has been conducted on early detection of smoke by performing image processing on an image captured by a surveillance camera.

  As an example, there is a conventional smoke detection device that detects smoke by installing a camera in a tunnel or the like and performing image processing on an image captured by the camera. In image processing for detecting smoke, generally, a reference image (reference image) is stored in advance, a difference image between the latest captured image and the reference image is calculated, and a region where a change has occurred is calculated. Extraction detects smoke. (For example, refer to Patent Document 1).

  In addition, the reference image is regularly updated in order to cope with the temporal change of the reference image due to the influence of sunlight.

Thus, by performing image processing on the image captured by the camera and performing smoke detection, the following two merits can be obtained.
1) By visually confirming the image of the surveillance camera, it is possible to grasp the smoke detection status in a remote place.
2) It is possible to divert already installed surveillance cameras and construct efficient equipment.

Japanese Patent No. 3909665

However, the prior art has the following problems.
It is preferable that the image of the surveillance camera is always taken in an environment where the illumination conditions and the visual field are stable. However, depending on the monitoring location, there are not only such favorable conditions, but there are places where moving objects such as people come and go within the monitoring range, and there are places where the sunshine conditions change with time, etc. It is conceivable that an area that is not suitable for monitoring is included in the monitoring range.

  In the prior art, smoke is detected based on the same criteria for areas that are not suitable for such monitoring, and smoke is generated due to the intrusion of moving objects and changes in sunlight conditions (effects of disturbance). May be erroneously detected. It is also possible to divide the image into areas of a predetermined size in a matrix and mask the areas (areas) not suitable for monitoring so that smoke detection is not performed, but the range that can be monitored is limited. The result will be.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to obtain a smoke detection device that suppresses erroneous detection due to the influence of disturbance without limiting the monitoring range.

  The smoke detection device according to the present invention includes a smoke feature amount calculation unit that extracts a feature amount related to smoke for each of a plurality of regions set in an image captured by a monitoring camera, and a plurality of regions. A predetermined reference determination value for determining the presence or absence of smoke generation is stored as a plurality of values having a plurality of detection sensitivities for each region, and a plurality of regions according to the monitoring target by the monitoring camera A sensitivity setting unit for each region for setting a desired detection sensitivity for each of the plurality of regions, and a sensitivity setting unit for each region from among a plurality of predetermined reference determination values stored in the storage unit in each of the plurality of regions. A predetermined reference determination value corresponding to the desired detection sensitivity is taken out, compared with the feature amount extracted by the smoke feature amount calculation means, and in each of the plurality of regions based on the comparison result In detection sensitivity of Nozomu those and a smoke judging means for detecting the occurrence of smoke.

  According to the smoke detection device of the present invention, depending on the monitoring target, by performing smoke detection using a reference determination value having a desired detection sensitivity for each region, the disturbance can be measured without limiting the monitoring range. Thus, it is possible to obtain a smoke detection device that suppresses erroneous detection due to the influence of the above.

  Hereinafter, a preferred embodiment of a smoke detection device of the present invention will be described with reference to the drawings.

Embodiment 1 FIG.
FIG. 1 is a configuration diagram of a smoke detection device according to Embodiment 1 of the present invention. The smoke detection apparatus according to the first embodiment includes an image memory 10, a storage unit 20, a smoke feature amount calculation unit 30, an area-specific sensitivity setting unit 40, and a smoke determination unit 50. The image memory 10 is configured as an image memory for a plurality of frames so that images captured by the camera 1 can be stored as time-series data for a certain period in the past.

  An image to be monitored is divided into a plurality of predetermined areas, and a plurality of areas are set in the image. FIG. 2 is an explanatory diagram relating to region segmentation of an image to be monitored in the first embodiment of the present invention. FIG. 2 illustrates an example in which an image to be monitored is divided in advance in a matrix shape, for example, into 20 areas of 4 × 5 in length. The storage unit 20 stores a plurality of values having different detection sensitivities in advance as reference determination values for determining the presence or absence of smoke generation for each of a plurality of regions.

  In the following description, in order to simplify the description, a case where the detection sensitivity is divided into three levels of high level, medium level, and low level will be exemplified. In this way, when the detection sensitivity is divided into three levels, the storage unit 20 stores and stores three different reference determination values for high level, medium level, and low level for each region. Has been.

  The smoke feature quantity calculation means 30 extracts a feature quantity related to smoke for each region of the captured image stored in the image memory 10, and determines whether or not there is a high possibility that smoke has occurred. The following four methods can be cited as typical feature value extraction methods.

[Extraction method 1: Smoke detection based on luminance distribution of pixels]
The smoke feature quantity calculation means 30 calculates the luminance variance of the pixels in each area for each area. As the image for calculating the luminance dispersion, the smoke feature amount calculation means 30 basically uses the latest captured image. However, by using time-series data of images stored in a plurality of image memories 10, a plurality of captured images can be used retroactively. In this way, the smoke feature quantity calculation means 30 outputs the calculated luminance dispersion or the standard deviation obtained from the brightness dispersion as a feature quantity relating to smoke.

  The smoke determination unit 50 described later determines whether or not the luminance dispersion calculated by the smoke feature amount calculation unit 30 or the standard deviation obtained from the luminance dispersion is within a predetermined range. When it is inside, it can be determined that there is a high possibility that smoke has been generated.

[Extraction method 2: Smoke detection based on temporal dispersion of average luminance of pixels]
The smoke feature quantity calculation means 30 calculates the average luminance of the pixels in each area for each area. Next, the smoke feature quantity calculation means 30 uses the time-series data of the images stored in the plurality of image memories 10 to go back to the past and average the plurality of captured images over a certain period in the same region. The brightness is calculated, and time series data of average brightness is generated for each target area. Then, the smoke feature amount calculating unit 30 calculates the luminance variance of the generated time series data of the average luminance.

  In this way, the smoke feature quantity calculating means 30 outputs the brightness variance calculated based on the time series data of the average brightness or the standard deviation obtained from the brightness variance as the feature quantity related to smoke.

  The smoke determination unit 50 described later determines whether or not the luminance dispersion calculated by the smoke feature amount calculation unit 30 or the standard deviation obtained from the luminance dispersion is within a predetermined range. When it is inside, it can be determined that there is a high possibility that smoke has been generated.

[Extraction method 3: Smoke detection based on low frequency intensity of average luminance of pixels]
The smoke feature amount calculation means 30 generates time series data of average luminance for each target area in the same manner as the extraction method 2 described above. Then, the smoke feature quantity calculation means 30 performs Fourier transform on the generated time series data of the average luminance to calculate a power spectrum.

  Next, the smoke feature quantity calculating means 30 extracts a predetermined low frequency component from the power spectrum calculated based on the time series data of the average luminance, and calculates the intensity for the mode. In this way, the smoke feature quantity calculation means 30 outputs the intensity of the low frequency component calculated based on the time series data of the average luminance as a feature quantity related to smoke.

  And the smoke determination means 50 mentioned later can judge that the possibility that smoke generate | occur | produced is high, when the intensity | strength calculated by the smoke feature-value calculation means 30 is below a predetermined value.

[Extraction method 4: Smoke detection based on difference average with reference image]
For each region, the smoke feature amount calculation means 30 obtains a luminance difference value between each pixel in the region and the corresponding pixel of the reference image stored in the image memory 10 in advance. Further, the smoke feature amount calculation means 30 calculates an average value of the luminance difference values for each region. In this way, the smoke feature quantity calculation means 30 outputs the average value of the luminance difference values as the feature quantity related to smoke.

  And the smoke determination means 50 mentioned later can judge that the possibility that smoke generate | occur | produced is high, when the average value calculated by the smoke feature-value calculation means 30 is larger than a predetermined value.

  Next, the sensitivity setting unit 40 for each region will be described. The area-specific sensitivity setting means 40 is a means for setting a desired detection sensitivity for each of a plurality of areas divided in advance. The operator or the like can set a desired area in a specific area by using the area-specific sensitivity setting means 40. The detection sensitivity can be set manually in advance. An image captured as a monitoring target does not necessarily require smoke detection with the same detection sensitivity in all areas. For example, depending on the monitoring target, set a higher detection sensitivity for areas where smoke is likely to occur, and lower the detection sensitivity for areas where there are many false alarm factors or where smoke is unlikely to occur. By setting to, it is possible to realize appropriate smoke detection that suppresses false detection and non-detection.

  Therefore, the area-specific sensitivity setting means 40 selects and sets a desired detection sensitivity from, for example, a high level, a medium level, and a low level for each area according to the monitoring target. I can keep it.

  FIG. 3 is an illustration of desired detection sensitivity set for each region of the image to be monitored in the first embodiment of the present invention. FIG. 3A shows a case where all the detection sensitivities of the 20 divided areas are set as “medium level”. On the other hand, FIG. 3B shows a case where the detection sensitivity is set by dividing into “high level”, “medium level”, and “low level” according to the region. In this way, by setting a desired detection sensitivity for each region in accordance with the image to be monitored, it is possible to realize appropriate smoke detection that suppresses erroneous detection and non-detection. As a method of setting sensitivity, considering that smoke flows upward, high sensitivity may be set in the upper area of the image, and low sensitivity may be set in the lower part of the image in which human movement is likely to appear. Further, the vicinity of illumination that causes a change in luminance and causes disturbance may be set to low sensitivity.

  Next, in each of the plurality of areas, the smoke determination unit 50 determines the sensitivity for each area from among the reference determination values corresponding to the three detection sensitivities of high level, medium level, and low level stored in the storage unit 20. A reference determination value corresponding to the desired detection sensitivity set by the setting means 40 is taken out. Furthermore, the smoke determination unit 50 compares the feature amount extracted by the smoke feature amount calculation unit 30 with the extracted reference determination value, and determines whether or not smoke is highly likely based on the comparison result. Judging. When the reference value serving as the threshold is a predetermined range, the high level predetermined range is narrower than the medium level predetermined range, and the medium level predetermined range is a small level predetermined range. It is set narrower than the range.

  Thereby, the smoke determination means 50 can detect the generation of smoke with a desired detection sensitivity in each of the plurality of regions.

  When the specific extraction methods 1 to 4 of the smoke feature amount calculating unit 30 are described as an example, the reference determination values stored in advance in the storage unit 20 are as follows. When extraction method 1 is used for extraction of feature quantities related to smoke, three detection sensitivities of high level, medium level, and low level are supported as reference judgment values for luminance variance or standard deviation obtained from luminance variance. The determined reference determination value is stored in the storage unit 20.

  In addition, when the extraction method 2 is used for extracting the feature quantity related to smoke, the luminance variance calculated based on the time series data of the average luminance or the standard determination value of the standard deviation obtained from the luminance variance is high. The reference determination values corresponding to the three detection sensitivities of level, medium level, and low level are stored in the storage unit 20.

  Further, when the extraction method 3 is used for extraction of the feature amount related to smoke, the reference determination value of the intensity of the low frequency component calculated based on the time series data of the average luminance is high level, medium level, low level. The reference determination values corresponding to the three detection sensitivities are stored in the storage unit 20.

  In addition, when the extraction method 4 is used for extracting the feature quantity regarding smoke, the reference determination value corresponding to the three detection sensitivities of high level, medium level, and low level is used as the reference determination value of the average value of the luminance difference values. Is stored in the storage unit 20.

  As described above, according to the first embodiment, the monitoring area is divided into a plurality of matrix-like areas, and the detection sensitivity is set in each small area in consideration of the possibility of smoke generation. it can. Therefore, smoke detection can be performed using a reference determination value having a desired detection sensitivity for each region in accordance with the monitoring target. As a result, it is possible to realize a smoke detection device that suppresses erroneous detection due to the influence of disturbance without limiting the monitorable range.

  In addition, it is not necessary to select and use any one of the extraction methods 1 to 4 exemplified as specific extraction methods of the feature quantity regarding smoke. The smoke determination means 50 can finally determine whether smoke is generated from the determination results obtained by the plurality of extraction methods. In this case, the reference determination values corresponding to the plurality of sensitivity levels are further stored in the storage unit 20 corresponding to the plurality of extraction methods.

Embodiment 2. FIG.
In the first embodiment, a case has been described in which a desired detection sensitivity is manually set in advance for each region according to a monitoring target (so-called static setting). On the other hand, in the second embodiment, a case where the presence or absence of the influence of a disturbance such as a moving object or illumination change is determined based on the analysis result of the current captured image, and the detection sensitivity is dynamically changed will be described. To do.

  FIG. 4 is a configuration diagram of the smoke detection apparatus according to Embodiment 2 of the present invention. Compared with the configuration of FIG. 1 in the first embodiment, the configuration of FIG. 4 in the second embodiment is different in that a disturbance occurrence detecting means 60 is newly provided. Therefore, the function of the disturbance occurrence detecting means 60 will be mainly described below.

  The image memory 10 is configured as an image memory for a plurality of frames so that images captured by the camera 1 can be stored as time-series data for a certain period in the past. Then, the disturbance occurrence detection means 60 uses the time series data stored in the image memory 10 to calculate a feature amount based on a temporal change in luminance in each of a plurality of previously divided areas. This feature amount is an index for determining (identifying) whether or not an influence of disturbance due to a moving object or a change in illumination occurs, and a specific example thereof will be described later.

  Further, the disturbance occurrence detection unit 60 determines whether or not an influence of disturbance has occurred based on the comparison result between the calculated feature amount and the predetermined reference value, and transmits the determination result to the smoke determination unit 50.

  On the other hand, in the area where the influence of the disturbance is determined to be generated by the disturbance occurrence detecting means 60, the smoke determination means 50 uses one reference from a plurality of predetermined reference determination values stored in the storage unit 20. When the determination value is extracted, a reference determination value having a detection sensitivity lower than the desired detection sensitivity set by the area-specific sensitivity setting means 40 is extracted. As a result, the smoke determination means 50 detects the generation of smoke with a detection sensitivity lower than the desired detection sensitivity in the region where the influence of the disturbance is determined to occur.

  For example, it is assumed that a region in which the desired detection sensitivity is set to a medium level by the region-specific sensitivity setting unit 40 is determined by the disturbance occurrence detection unit 60 as a region where the influence of the disturbance is generated. In this case, in order to detect the generation of smoke in this region, the smoke determination means 50 takes out the low-level reference determination value one rank below from the storage unit 20 instead of the intermediate level that is the desired reference determination value. Thus, the smoke determination is performed based on the comparison with the feature amount extracted by the smoke feature amount calculation means 30.

  Further, for example, it is assumed that a region where the desired detection sensitivity is set to a high level by the region-specific sensitivity setting unit 40 is determined as a region where the influence of the disturbance is generated by the disturbance generation detection unit 60. In this case, in order to detect the generation of smoke in this region, the smoke determination means 50 is not a high reference level that is a desired reference determination value, but a medium level reference determination value that is one rank below, or two ranks below it. The low-level reference determination value is taken out from the storage unit 20, and the smoke determination is performed based on the comparison with the feature amount extracted by the smoke feature amount calculation means 30.

  Here, when lowering the reference determination value, it can be defined as follows whether the rank is lowered by one rank from the high level to the middle level or by two ranks from the high level to the low level. First, when it is determined that the influence of the disturbance is generated by the disturbance occurrence detection means 60, a rule that only lowers the rank from the desired detection level or lowers it to a low level in advance. It is possible to prescribe.

  Second, it is conceivable that the disturbance occurrence detection means 60 has two predetermined reference values used for comparison, and defines whether the rank is lowered by one rank or two ranks according to the comparison result. In addition, when it is determined that a disturbance has occurred in the adjacent region, the sensitivity may be lowered to a low sensitivity. For example, the sensitivity of its own area may be lowered according to the number of areas in which disturbances occur in the eight neighboring areas around the area to be monitored.

  Next, a specific example of a feature amount that is an index for determining whether or not an influence of disturbance has occurred will be described. Here, two factors will be described in detail, which are (1) the case where the luminance is changed due to a change in illumination and (2) the case where the luminance is changed due to a moving object such as a passerby. In either case, it is important to distinguish from a change in luminance due to the generation of smoke.

(1) When the luminance changes due to a change in illumination Both the luminance change due to the illumination change and the luminance change due to the smoke are common in that both slowly and gradually change. Therefore, it is difficult to distinguish both from the amount of change in luminance. However, if attention is paid to the transition of the luminance change, the two can be distinguished. FIG. 5 is a diagram showing the transition of the change amount of the average luminance in Embodiment 2 of the present invention, where the vertical axis is the change amount of the average luminance value, and the horizontal axis is the number of cycles.

  As shown in FIG. 5, the change in luminance due to the change in illumination tends to change continuously (that is, monotonously increase or monotonously decrease) as a whole affected by the illumination. On the other hand, the change in luminance due to the effect of smoke tends to change while oscillating without exhibiting a constant change in the amount of change in average luminance due to the characteristics of the smoke itself.

  Therefore, the disturbance occurrence detection means 60 calculates the average luminance for each region with respect to the time-series data related to the monitoring target image, and counts +1 when the average luminance value changes in the increasing direction. When the value changes in a decreasing direction, -1 is counted. Then, the disturbance occurrence detecting means 60 counts the change in the average brightness in a cycle finer than the vibration of the change amount of the average brightness due to the influence of smoke.

  When counting is performed in this manner, when the average luminance changes due to a change in illumination, the count value monotonously increases (or monotonously decreases), and reaches a predetermined count value after a certain period of time has elapsed. On the other hand, when the average luminance changes due to the influence of smoke, the count value repeatedly increases and decreases, and does not reach the predetermined count value even after a certain period of time.

  Therefore, the disturbance occurrence detection means 60 clearly counts the change in luminance due to the change in illumination by counting the change in average luminance value for each divided area with respect to the time-series data related to the image to be monitored. Can be identified. Therefore, when the disturbance occurrence detection unit 60 determines that a change in luminance due to a change in illumination has occurred, the disturbance generation detection unit 60 outputs that effect to the smoke determination unit 50, assuming that the influence of the disturbance has occurred.

  On the other hand, when the disturbance occurrence detection unit 60 detects a change in luminance due to a change in illumination, the smoke determination unit 50 lowers the determination reference value and performs smoke detection. As a result, the smoke determination means 50 can detect the generation of smoke with a detection sensitivity lower than the desired detection sensitivity in an area where it is determined that an influence of disturbance (a change in luminance due to a change in illumination) has occurred. It is possible to prevent erroneous detection when the detection sensitivity is not changed, or undetected by masking.

(2) When the brightness changes due to a moving object such as a passerby In order to distinguish between a change in brightness due to a moving object such as a passerby and a change in brightness due to smoke, the correlation value between temporally changing images Pay attention to. In a region where the brightness is blurred due to the generation of smoke, the correlation value tends to gradually decrease or increase due to the characteristics of the smoke itself. On the other hand, in a region where a normal background is shielded by a moving object such as a passerby, the correlation value tends to change rapidly.

  Therefore, the disturbance occurrence detection means 60 calculates, for example, the average luminance of each area of the image one cycle before and the current, and calculates the correlation value of the same area. Then, the disturbance occurrence detecting means 60 sequentially obtains such correlation values with respect to the time-series data related to the monitoring target image, and determines whether or not the amount of change in the correlation values is equal to or greater than a predetermined value. It is possible to clearly identify a change in luminance due to a moving object such as a passerby. Therefore, when the disturbance occurrence detection unit 60 determines that a change in luminance due to a moving object such as a passerby has occurred, the disturbance determination unit 60 determines that the influence of the disturbance has occurred, and notifies the smoke determination unit 50 to that effect. Output.

  On the other hand, when the disturbance occurrence detection unit 60 detects a change in luminance due to a moving object such as a passerby, the smoke determination unit 50 lowers the determination reference value and performs smoke detection. As a result, the smoke determination means 50 generates smoke with a detection sensitivity lower than the desired detection sensitivity in an area where it is determined that an influence of disturbance (change in luminance due to a moving object such as a passerby) has occurred. It is possible to detect and prevent erroneous detection when the detection sensitivity is not changed or undetected by masking.

  It should be noted that the area for obtaining the feature value by the disturbance occurrence detection means 60 does not have to be the same area as a plurality of areas divided in advance as areas for smoke detection, and can be set separately.

  As described above, according to the second embodiment, smoke detection can be performed using the reference determination value having a desired detection sensitivity for each region according to the monitoring target. In addition, based on the time-series data related to the image to be monitored, a feature value indicating that the influence of the disturbance has occurred is obtained, and if it is determined that the influence of the disturbance has occurred, the detection sensitivity is dynamically changed. Can do. As a result, it is possible to realize a smoke detection device that suppresses erroneous detection due to the influence of disturbance without limiting the monitorable range. If the disturbance occurrence detection unit determines that the influence of the disturbance is a very small region, the medium sensitivity may be switched to the high sensitivity side.

  In particular, when the brightness changes due to a change in illumination, or when the brightness changes due to a moving object such as a passerby, it is not a steady disturbance but is affected by a disturbance that changes over time. It is effective in some cases.

  In the above description, the case has been described in which the detection sensitivity is lowered when it is determined that the influence of disturbance has occurred. On the other hand, if it is determined that the influence of the disturbance has disappeared after the detection sensitivity is once lowered, the detection sensitivity can be restored to the desired detection sensitivity by returning the detection sensitivity to the original one. In each of the embodiments, the sensitivity setting has been described in three stages of high, medium, and low. However, the number of stages of sensitivity may be switched between two stages or four or more stages, for example. Further, the number of steps of sensitivity setting may be determined in accordance with how to calculate the smoke feature amount.

It is a block diagram of the smoke detection apparatus in Embodiment 1 of this invention. It is explanatory drawing regarding the area division | segmentation of the image used as the monitoring object in Embodiment 1 of this invention. It is an illustration figure of the desired detection sensitivity set with respect to each area | region of the image used as the monitoring object in Embodiment 1 of this invention. It is a block diagram of the smoke detection apparatus in Embodiment 2 of this invention. It is the figure which showed transition of the luminance variation in Embodiment 2 of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Camera, 10 Image memory, 20 Memory | storage part, 30 Smoke feature-value calculation means, 40 Sensitivity setting means according to area | region, 50 Smoke determination means, 60 Disturbance generation | occurrence | production detection means

Claims (4)

Smoke feature quantity calculating means for extracting a feature quantity related to smoke for each of a plurality of regions set in an image captured by the monitoring camera;
In each of the plurality of regions, a predetermined reference determination value for determining the presence or absence of smoke generation is stored as a plurality of values having a plurality of detection sensitivities for each region,
Sensitivity setting means for each region that performs desired detection sensitivity setting for each of the plurality of regions according to the monitoring target by the monitoring camera;
In each of the plurality of regions, a predetermined reference determination value corresponding to the desired detection sensitivity set by the region-specific sensitivity setting means is selected from a plurality of predetermined reference determination values stored in the storage unit. A smoke determination unit that detects the generation of smoke with the desired detection sensitivity in each of the plurality of regions based on a comparison result, and is extracted and compared with the feature amount extracted by the smoke feature amount calculation unit. Smoke detection device characterized by. The smoke detection device according to claim 1,
In order to identify whether the temporal change in luminance in each of the plurality of regions is due to the influence of disturbances other than the generation of smoke, based on time-series data composed of a plurality of images captured in time series by the monitoring camera A disturbance occurrence detecting means for calculating whether or not the influence of the disturbance has occurred based on a comparison result between the feature quantity and a predetermined reference value,
In the area where the influence of the disturbance is determined to be generated by the disturbance occurrence detecting means, the smoke determination means is one reference determination from among a plurality of predetermined reference determination values stored in the storage unit. When the value is extracted, a reference determination value having a detection sensitivity lower than the desired detection sensitivity set by the area-specific sensitivity setting means is extracted, and smoke generation is detected with a detection sensitivity lower than the desired detection sensitivity. A smoke detection device characterized by: The smoke detection device according to claim 2,
The disturbance occurrence detecting means calculates an average luminance change amount in each of a plurality of regions as the feature amount based on the time series data, and the average luminance change amount monotonously increases or monotonously decreases to a predetermined amount. Therefore, it is determined that an illumination change has occurred as an influence of the disturbance. The smoke detection device according to claim 2,
The disturbance occurrence detecting means sequentially calculates a correlation value of luminance between one cycle before and the present as the feature value based on the time series data, and the change amount of the correlation value is equal to or greater than a predetermined value The smoke detection device according to claim 1, wherein it is determined that a temporary shielding by a moving object has occurred as an influence of the disturbance.

Images