JP2003109012A - Device for detecting abnormality - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an abnormality detecting device capable of detecting whether or not an object to be monitored enters a room and whether or not the object leaves the room on the basis of an output of an imaging means provided to detect abnormality of the object. SOLUTION: This abnormality detecting device is provided with a room entering detecting means for detecting that the object to be monitored enters the room on the basis of decision results made by a first movement deciding means and decision results made by a second movement deciding means, a means for setting up a room entering state flag when the room entering detecting means detects that the object enters the room, a room leaving detecting means for detecting that the object leaves the room on the basis of decision results of the first and second movement deciding means, and a means fro resetting the room entering state flag when the room leaving detecting means detects that the object leaves the room.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an abnormality detecting device for detecting an abnormality such as a person falling in a room such as a bathroom or a toilet.

[0002]

2. Description of the Related Art Accidents of bathers dying during bathing are increasing year by year in the bathroom, and the number of bathing fatalities during bathing exceeds the number of fatalities due to traffic accidents. Is getting attention.

A direct cause of an accident while taking a bath is that the patient may fall down due to heart failure or stroke while taking a bath, or consciousness may be impaired in the bathtub, resulting in drowning. However, although the cause of such a state is being investigated, the current situation is that it is difficult to identify the factors due to the current changes in lifestyle. Therefore, it is not easy to prevent heart failure or stroke while taking a bath.

However, even if it is not possible to prevent the occurrence of symptoms, there is a high possibility that a person will be killed by early detection.

[0005]

Therefore, the applicant of the present invention is
We have developed the following abnormality detection device. That is, the image in the bathroom is captured by the image capturing means. The image captured by the image capturing unit is divided into a plurality of feature amount calculation regions, and the image feature amount is extracted from the image captured by the image capturing unit for each feature amount calculation region.

The amount of temporal change in the image feature amount extracted for each feature amount calculation area is detected at predetermined time intervals. The temporal change amount of the image feature amount detected for each feature amount calculation region and the motion determination threshold value are compared for each feature amount calculation region, and the temporal change amount of the image feature amount is equal to or less than the motion determination threshold value. If there is a motion, it is determined that there is no motion in the feature amount calculation region, and if it is greater than the motion determination threshold value, it is determined that there is motion in the feature amount calculation region.

When the number of characteristic amount calculation areas determined to be moving is less than or equal to the reference value, it is determined that there is no movement in the bathroom, and the number of characteristic amount calculation areas determined to be motion is greater than the reference value. It is determined that there is movement in the bathroom. Then, it is determined whether or not an abnormality has occurred in the bather based on the result of the movement determination in the bathroom a predetermined number of times in the past.

By the way, the above-mentioned abnormality detecting device determines that an abnormality has occurred when the bather does not move for a predetermined period of time or longer, and therefore when the bather is in the bathroom. , It is necessary to judge abnormality.

Whether or not a bather has entered the bathroom and whether or not the bather has left the bathroom has conventionally been detected based on on / off of a lighting switch of an illumination lamp in the bathroom. However, in the conventional method, there is a problem that the lighting switch of the illumination lamp in the bathroom and the abnormality detection device must be connected, and when the bather enters the bathroom without turning on the illumination lamp in the bathroom. Has a problem that it is not possible to detect the abnormality of the bather.

According to the present invention, whether or not the monitored object has entered the room and whether or not the monitored object has left the room is based on the output of the image pickup means provided for detecting the abnormality of the monitored object. It is an object of the present invention to provide an abnormality detection device that can detect an abnormality.

Further, according to the present invention, whether or not the monitoring target has entered the room and whether the monitoring target has left the room are output to the output of the image pickup means provided for detecting the abnormality of the monitoring target. An object of the present invention is to provide an abnormality detection device that can be detected based on the above, and can determine whether or not the monitoring target is abnormal only when the monitoring target is in the room.

[0012]

An abnormality detecting apparatus according to the present invention includes an image pickup means for picking up an image of a room, a video image picked up by the image pickup means, divided into a plurality of feature amount calculation regions, and each feature amount calculation region is divided into a plurality of feature amount calculation regions. Extraction means for extracting the image feature amount from the video imaged by the image pickup means, detection means for detecting the temporal change amount of the image feature amount extracted for each feature amount calculation region, and each feature amount calculation region The temporal change amount of the image feature amount detected for each and the motion determination threshold value are compared for each feature amount calculation region, and if the temporal change amount of the image feature amount is less than or equal to the motion determination threshold value, the feature is calculated. A first motion determination unit that determines that there is no motion in the amount calculation region, and determines that there is motion in the feature amount calculation region when it is larger than the motion determination threshold value, the number of feature amount calculation regions that is determined to be motion is the reference There is no movement in the room when it is below the value And a determination result by the first motion determination means and a second motion determination that determines that there is motion in the room when the number of feature amount calculation regions that is determined to be motion is larger than the reference value. Based on the determination result by the means, the room entrance detection means for detecting that the monitoring target has entered the room, and the room entrance state flag is set when the room entrance detecting means detects that the monitoring object has entered the room. That the monitored object has left the room by the leaving detection means and the leaving detection means for detecting that the monitored object has exited the room based on the determination result by the first motion determination means and the second motion determination means Means for resetting the entry state flag when the detection is made, and if the entry state flag is set, Zui it, characterized in that it comprises an abnormality detecting means for detecting a monitored abnormal.

As the entry detecting means, for example, when the entry state flag is reset, the second movement determining means determines that there is movement in the room, and the first movement determining means determines the entire image feature amount calculation area. After it is determined that there is movement in the entry determination region set in the peripheral portion of the inside of the room, after the first predetermined time or more has passed and before the second predetermined time has passed, the second movement determination means When it is determined that there is movement in the room and it is determined by the first movement determination means that there is movement in the image feature amount calculation area in the central portion of the entire image feature amount calculation area, the monitoring target enters the room. What is determined to have been used is used.

The entry determination area may be fixed,
The determination may be made based on the history of the motion determination result by the first motion determining means at the time of the past entry detection.

As the exit detection means, for example, when the entry state flag is set, the second movement determination means determines that there is movement in the room, and the first movement determination means.
After the third predetermined time has elapsed after the movement determining means determines that there is movement in the central portion of the entire image feature amount calculation area, the second movement determining means determines that there is movement in the room and 1 When the movement determination means determines that there is movement in the entry determination area, and the second movement determination means does not determine that there is movement in the room until the fourth predetermined time has elapsed, the monitoring target is The one that is judged to have left the room is used.

As the abnormality detecting means, for example, when the entry state flag is set, based on the result of the predetermined number of past motion determinations by the second motion determining means,
A device that determines whether or not an abnormality has occurred in the monitoring target in the room is used.

The motion determination threshold value is set to, for example, the first threshold value when the entry state flag is reset, and the second threshold value which is smaller than the first threshold value when the entry state flag is set. It is set to a threshold.

[0018]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a schematic view showing an arrangement example of an abnormal state detecting device.

In FIG. 1, 1 is a bathroom. 2 is a dressing room. 3 is a corridor. 4 is a kitchen.

A door 5 is provided on the wall between the bathroom 1 and the dressing room 2.
Is opened and closed so that a person can enter the bathroom 1 from the dressing room 2 or leave the dressing room 2 from the bathroom 1 by opening the door 5. On the wall between the bathroom 1 and the dressing room 2 on the dressing room 2 side, a lighting switch 6 for turning on and off an illumination lamp 7 grounded in the bathroom 1 is arranged.

Reference numeral 9 is an abnormal condition detecting device for detecting that the bather is in an abnormal condition. More specifically, the abnormal state detection device 9 has a built-in camera 8 (see FIG. 2) that captures an image of the inside of the bathroom 1 from above the bathroom, and the bather is in an abnormal state based on the image of the camera 8. It detects that it became.

Reference numeral 10 is a first alarm device arranged in the bathroom 1 to give an alarm to a person (bather) who is in the bathroom 1. Reference numeral 11 denotes a second alarm device arranged in the kitchen 4, for example, to notify a person outside the bathroom 1 of the abnormality of the bather.

FIG. 2 shows the electrical construction of the abnormal state detecting device 9.

The camera 8 comprises a lens 8a and a CCD 8b. The CCD 8b photoelectrically converts the optical image in the bathroom through the lens 8a and outputs it as an electric signal. CCD8
The output signal of b is sent to the preprocessing circuit 91. The preprocessing circuit includes a CDS circuit, an AGC circuit and an ADC.

The output signal of the CCD 8b is subjected to correlated double sampling processing in the CDS circuit, further gain adjusted to an optimum amplitude in the AGC circuit, converted into a digital signal by the ADC, and input to the YC separation circuit 92. .

The YC separation circuit 92 generates a luminance signal Y and color difference signals RY and BY from the digital CCD signal.
The signals Y, RY, generated by the YC separation circuit 92,
BY is sent to the integration circuit 93 for each area.

As shown in FIG. 3, the region-by-region integrating circuit 93 has a plurality of regions having the same area (feature amount calculation region) on one screen.
E11 to E88 are divided, and in each feature amount calculation area,
Integration processing or averaging processing is performed for each of the signals Y, RY, and BY. In this example, the integration process is performed for each of the signals Y, RY, and BY. The result (hereinafter, referred to as image feature amount) accumulated for each of the regions E11 to E88 is the image feature amount R.
It is stored in AM96.

The timing controller 99 outputs a CCD drive pulse for driving the CCD 8b. The timing control unit 99 also outputs a sampling pulse for the CDS circuit and a sampling clock for the ADC.

Timing control is performed so that predetermined YC separation, integration for each area, integration of high frequency components, and writing to the image feature RAM are normally performed with respect to the CCD digital signal input at a constant timing by the timing control unit 99. The unit 99 sends a control signal to each circuit. Further, the timing control unit 99 outputs a Vsync signal, which is a vertical synchronizing signal of a video, as an interrupt signal to a microcomputer (hereinafter, microcomputer) 97 that makes a motion determination.

The microcomputer 97 sets parameters such as position setting of the area to the timing control unit 99.
In addition, the microcomputer 97 controls the gain of the AGC circuit and the timing control through the CPU I / F 98 in order to optimize the brightness of the image, and controls the exposure time of the CCD 8b. Further, the microcomputer 97 uses the timing control unit 9
An interrupt signal is received from 9 and motion determination is performed for each feature amount calculation region in synchronization with the output of video. Further, the microcomputer 97 performs entry determination, exit determination, abnormality determination, etc. based on the motion determination result.

The central 4 × 4 region in the entire feature amount calculation region S0 including the 8 × 8 feature amount calculation region shown in FIG. 4 is referred to as a central region S1. Further, the outermost portion in the total feature amount calculation area S0 is referred to as a peripheral area S2.

As shown in FIGS. 5A to 5D, 16 shaded areas R1 to R16 set in the peripheral area S2.
Will be referred to as entry / exit area candidates.

6 and 7 show the operation of the microcomputer 97.

The flags and variables used in the motion determination process will be described.・ Bathing flag: This is a flag that stores whether or not a bather is in the bathroom. When the bather is not in the bathroom (waiting for entry), the flag is 0, and the bather is in the bathroom. It becomes 1 in the case (entry state).・ Temporary entry flag: This is a flag that remembers whether or not there is a possibility that a bather may have entered the bathroom. If the bather has no possibility of entering the bathroom, it will be 0, and the bather has entered the bathroom. It is 1 when there is a possibility. -Exit flag: A flag that remembers whether or not a bather may have left the bathroom. It is 0 if there is no possibility that the bather has left the bathroom, and the bather may have left the bathroom. It becomes 1 when there is a property.・ Number of room entrances: The number of times a bather entered the bathroom.

When the power of the microcomputer 97 is turned on, the microcomputer 97 performs initial setting (step 1). In this initial setting, the various flags and the number of times of entering the room are set to 0, and the temporary entry elapsed time, which will be described later, is set to 0. Also,
In the entry waiting state, a large value SUBTH_1 is set as the threshold value SUBTH used when determining the presence or absence of motion in order to reduce the motion detection sensitivity and reduce the erroneous determination of motion due to changes in outside light. As will be described later, in the entry state, SUBTH_2 smaller than SUBTH_1 is set as the threshold SUBTH in order to increase the motion detection sensitivity.

When the initial setting is completed, it waits for the Vsync signal to be input (step 2). When the Vsync signal is input, the following processing is performed.

When the Vsync signal is input, the microcomputer 97
Determines the presence or absence of image movement for each feature amount calculation area, stores the feature amount calculation area determined to have image movement, and stores the number of feature amount calculation areas determined to have image movement. Is calculated (step 3).

That is, the image feature amount of each feature amount calculation area of the previous field stored in the image feature amount RAM 96 is read out through the CPU I / F 98 and stored in the RAM inside the microcomputer 97. When the image feature amount of each feature amount calculation area is stored in the RAM inside the microcomputer 97, the microcomputer 9
7, the variable CountM representing the number of areas with movement is set to 0.

Then, based on the amount of change in the feature amount of each feature amount calculation region, the motion determination for each feature amount calculation region is performed. That is, the microcomputer 97 uses the image feature amount RAM for each calculation area.
The absolute value | Dxy (t) −Dxy (t-1) | of the difference between the image feature amount Dxy (t) read this time from 96 and the image feature amount Dxy (t-1) previously read from the image feature amount RAM 96 is It is determined whether it is larger than the threshold value SUBTH.

This determination is made for each feature amount calculation area.
The change amount of the integrated value of the luminance signal Y, the change amount of the integrated value of the color difference signal R-Y, and the change amount of the integrated value of the color difference signal B-Y are performed.
If it is larger than H, it is determined that there is motion in the feature amount calculation area. Then, the motion determination result for each feature amount calculation area is stored.

Further, each time the microcomputer 97 determines that there is a motion in the characteristic amount calculation area, the microcomputer 97 increments the variable CountM by 1. In this way, when the motion determination for all the feature amount calculation areas is completed, the microcomputer 97
Indicates that the number of areas CountM determined to be moving is the reference value TH_
It is determined whether it is larger than Count (step 4). The reference value TH_Count is, for example, 1 of the total number of feature amount calculation areas.
It is set to about 50%. For example, when the total number of feature amount calculation areas is 64 as shown in FIG. 3, the reference value TH_Count is set to 7.

Then, the number of areas Co determined to have a motion Co
If untM is less than or equal to the reference value TH_Count, it is determined that there is no movement in the bathroom (step 5), and it is determined whether the entry flag is 0 (waiting for entry) (step 6). . If the entry flag is 0, it is determined whether the provisional entry flag is 1 (step 7). If the temporary entry flag is not 1, the process returns to step 2.

That is, when the bather is not in the bathroom, the processes of steps 2 to 7 are repeated.

When it is determined in step 4 that the number of areas CountM determined to be moving is larger than the reference value TH_Count, it is determined that there is movement in the bathroom (step 9), and the entry flag is 0. It is determined whether or not (the entry waiting state) (step 10). If the entry flag is 0, it is determined whether the provisional entry flag is 0 (step 11).

If the temporary entry flag is 0, it is determined whether the number of entry is equal to or more than a predetermined number of times A (step 12). When the number of times of entering the room is less than the predetermined number of times A, the entire peripheral area S2 is set as the entry determination area (step 13), and then the process proceeds to step 15. If the number of times of entering the room is larger than the predetermined number of times A, the entry area determined from the history of the temporary entry area (the entry area determined in step 24 described later) is set as the entry determination area (step 14). ), And proceeds to step 15.

In step 15, based on the processing result in step 3 this time, it is determined whether or not there is a movement in the entry determination area. If there is no movement in the entry determination area, it is determined that the determination result that there is movement in step 9 this time is not due to the bather entering the room, and the process returns to step 2.

If it is determined in step 15 that there is movement within the entry determination area, it is possible that a bather has entered the bathroom, so the temporary entry flag is set to 1 (step 16). Further, entry / exit area candidates R1 to R16
Of these, the region having the largest number of regions (the number of feature amount calculation regions) determined to be moving in this step 3 is set as the temporary entry region (step 17).

Then, the elapsed time of temporary entry is counted up (+1) (step 18). Then, it is determined whether or not the temporary entry elapsed time is equal to or less than a predetermined C seconds (step 19). If the temporary entry elapsed time is C seconds or less, it is determined whether the temporary entry elapsed time is a predetermined D seconds or more shorter than C seconds (step 20).

When the temporary entry elapsed time is less than D seconds, the process returns to step 2. When it is determined in step 4 that there is movement in the next field, the process proceeds to step 18 through step 9, step 10 and step 11, and the provisional entry elapsed time is counted up. Alternatively, when it is determined that there is no movement in step 4 in the next field, the process proceeds to step 8 through step 5, step 6 and step 7, and the provisional entry elapsed time is counted up.

Steps 2, 3, 4, 9, 10, 11, 1
When the temporary entry elapsed time is C seconds or less and D seconds or more at the time of shifting to step 19 on the route of 8,
Based on the processing result of step 3 this time, the central region S
It is determined whether or not there is movement within 1 (step 2
1).

If there is no movement within the central area S1, the process returns to step 2. Also in this case, when it is determined that there is movement in step 4 in the next field, the process proceeds to step 18 through step 9, step 10 and step 11, and the temporary entry elapsed time is counted up. Alternatively, if it is determined that there is no movement in step 4 in the next field, step 5,
Proceeding to step 8 through step 6 and step 7, the temporary entry elapsed time is counted up.

If it is determined in step 21 that there is movement within the central area S1, the entry flag is set to 1
At the same time as entering the room, the number of times of entering the room is incremented (+1) (step 22). That is, when a movement is detected in the entry determination area, and after a lapse of D seconds or more and a time of C seconds or less, when a movement is detected in the central area S1, a bather is bathed in the bathroom. It is determined that you have entered the room.

The history of the temporary entry area is updated using the temporary entry area set in step 17 (step 23). In addition, based on the history of the temporary entry area, the area candidate having the largest number of times of being determined to be the temporary entry area among the entry and exit area candidates R1 to R16 is determined as the entry area (step 24). Also, to increase the sensitivity of motion detection,
SUBTH_2 is set as the threshold value SUBTH (step 25). Then, the process returns to step 2.

In step 19, if the temporary entry elapsed time exceeds C seconds, that is, after D seconds or more have elapsed since the movement was detected in the entry determination area, until C seconds have elapsed, When no movement is detected in the central area S1, it is not determined that the bather has entered the bathroom. In this case, after the provisional entry elapsed time is set to 0 and the provisional entry flag is set to 0 (step 26), step 2
Return to.

In the room entering state (state where the room entering flag is 1), the process when it is determined in step 4 that there is movement will be described.

When it is determined in step 4 that there is movement in the room entering state, the process proceeds to step 27 through steps 9 and 10.

In step 27, it is determined whether or not there is a movement in the peripheral area S2 based on the processing result of step 3 this time. If there is no movement in the peripheral area S2, no exit area is set as the exit area (step 2).
8). In addition, based on the processing result of step 3 this time,
It is determined whether or not there is movement in the central area S1, and the determination result is stored as history information of the static / moving determination result in the central area (step 29). Then, after the exit flag is set to 0 (step 30), the process returns to step 2.

If it is determined in step 27 that there is movement in the peripheral area S2, the entry / exit area candidates R1 to R1 are selected.
Of R16, the area having the largest number of areas (the number of characteristic amount calculation areas) determined to be moving in step 3 of this time is set as the leaving area (step 31). Further, based on the processing result of step 3 this time, it is determined whether or not there is movement in the central region S1, and the determination result is stored as history information of the static / moving determination result in the central region (step 3
2). The history information of the static / moving judgment result in the central area is
It is also updated in step 35 described later. Then, it is determined whether or not motion is detected in the central area S1 within a predetermined G seconds in the past based on the history information of the static / moving determination result in the central area (step 33).

If no movement is detected in the central area S1 within the predetermined G seconds in the past, the exit flag is set to 0 (step 30), and the process returns to step 2. Past prescribed G
If the movement is detected in the central area S1 within the second, it is determined that the bather may have left the room, the exit flag is set to 1 (step 34), and the process returns to step 2.

In the room entering state (state where the room entering flag is 1), the process when it is determined in step 4 that there is no movement will be described.

When it is determined in step 4 that there is no movement in the entry state, the process proceeds to step 35 through steps 5 and 6. In step 35, stillness is stored as history information of the static / moving determination result in the central area.

Next, it is determined whether or not it has been determined that the bather has moved a predetermined number of M times or more in the past N (N field) movement determinations (motion determination in step 4) (step 36). If the bather has moved more than M times in the past N motion determinations, it is determined that the bather is in motion (step 37), and if not, it is determined that the bather is stationary. (Step 38).

When it is determined that the bather is still, it is determined whether or not the internal timer (not shown) is in the time counting operation (step 39), and if it is in the time counting operation, the process proceeds to step 41. To do. If the timekeeping operation is not in progress, start the timekeeping operation by the internal timer (step 4
0), the process proceeds to step 41.

After step 41, whether or not the time measured by the internal timer has exceeded the first predetermined time (step 41), the time measured by the internal timer has passed the second predetermined time shorter than the first predetermined time. Is determined (step 42), it is determined whether the time measured by the internal timer has passed a third predetermined time shorter than the second predetermined time (step 43).

When the time measured by the internal timer has not exceeded the third predetermined time (steps 41, 42, 43)
NO), and returns to step 2. When the time measured by the internal timer has not passed the second elapsed time but the third predetermined time has passed (NO in steps 41 and 42, YES in step 43), the entry area (determined in step 24) It is determined whether the latest entry area) is equal to the exit area (the latest exit area set in step 28 or 31) (step 44).

When the entry area and the exit area are the same,
It is determined whether or not the leaving flag is 1 (step 4
5). If the leaving flag is 1, it is determined that the bather has left the room, and the temporary entry flag, the entry flag, and the exit flag are set to 0, and the temporary entry elapsed time is set to 0 (step 46). As a result, the room enters the waiting state.

Further, in order to lower the motion detection sensitivity, SUBTH_1 is set as the threshold value SUBTH (step 47), and the process returns to step 2.

If it is determined in step 44 that the entry area and the exit area are not equal to each other or if it is determined in step 45 that the exit flag is 0, the first safety signal is output and the bathroom 1 is placed in the bathroom 1. The first alarm device 10 installed is driven to urge a person in the bathroom 1 to warn (step 50). And
Return to step 2.

If the time measured by the internal timer has not passed the first elapsed time but the second predetermined time has passed (NO in step 41, YES in step 42), the second
A first alarm device 10 that outputs a safety signal and is installed in the bathroom 1
Drive to warn people in bathroom 1 (step 4
9). The difference between the first safety signal and the second safety signal is that, for example, when the first alarm device 10 outputs an alarm sound,
The difference is that the warning sound is output in a small amount with the safety signal, and the warning sound is output in a large amount with the second safety signal. Alternatively, the difference between the first safety signal and the second safety signal is that, for example, when the first alarm device 10 outputs an alarm sound,
The difference is that the warning signal is output only once by the safety signal, and the warning sound is repeatedly output by the second safety signal.
Then, the process returns to step 2.

When the time measured by the internal timer has exceeded the first elapsed time (YES in step 41), the bathroom 1
It is determined that an abnormality has occurred in the bather inside, and an external alarm signal is output, and the second alarm device 1 installed outside the room, for example, in the kitchen 4
Drive 1 to notify an external person of the emergency (step 48). Then, the process returns to step 2.

The alarms in steps 49 and 50 are executed to wake up the bather who may be asleep in the bathroom 1. Also, step 4
The alarms at 9 and 50 are also executed to confirm the safety of the bather who is stationary in the bathtub or the like. The first alarm device 10 is provided to eliminate false alarms to the outside. The warning in step 48 determines that an abnormal state such as unconsciousness has occurred in the bathroom,
It notifies an external person of an emergency.

When it is determined in step 36 that the bather is moving (step 37), the timekeeping operation by the internal timer is reset (when the timekeeping operation is being performed) and the alarm signal is stopped. After (when the safety signal or the alarm signal is output) (step 51), the process returns to step 2.

In the above embodiment, the entry determination area is determined from the history of the temporary entry areas in the past. However, the entry determination area may be set in advance by the user and fixed to the set area. Good.

[0075]

According to the present invention, there is provided an image pickup means for detecting whether or not a monitoring target has entered the room and whether the monitoring target has left the room. It can be detected based on the output.

Further, according to the present invention, the image pickup means provided for detecting the abnormality of the monitoring target whether the monitoring target has entered the room and whether the monitoring target has left the room. It is possible to detect based on the output, and it becomes possible to determine whether or not the monitoring target is abnormal only when the monitoring target is in the room.

[Brief description of drawings]

FIG. 1 is a schematic diagram illustrating an arrangement example of an abnormal state detection device.

FIG. 2 is a block diagram showing a configuration of an abnormal state detection device.

FIG. 3 is a schematic diagram showing a plurality of feature amount calculation areas set in one screen.

FIG. 4 is a schematic diagram showing a central area S1 and a peripheral area S2 that are set inside the total feature amount calculation area S0.

5 is a schematic diagram showing 16 entry / exit area candidates R1 to R16 set in a peripheral area S2 of FIG. 4. FIG.

FIG. 6 is a flowchart showing the operation of the microcomputer.

FIG. 7 is a flowchart showing the operation of the microcomputer.

[Explanation of symbols]

8 cameras 8a lens 8b CCD 9 Abnormal condition detector 91 Preprocessing circuit 92 YC separation circuit 93 Integration circuit for each area 96 image feature RAM 97 Microcomputer 98 CPU I / F 99 Timing control unit

   ─────────────────────────────────────────────────── ─── Continued front page    F term (reference) 5B057 AA19 BA02 DA15 DB02 DB06                       DB09 DC22 DC25 DC32                 5C054 AA01 AA05 CA04 CC02 CE01                       CE16 CG07 CH02 EA01 FA09                       FC13 HA37                 5C086 AA22 BA04 CA28 CB36 DA08                       DA33 EA45                 5L096 AA02 AA06 BA02 CA02 DA03                       FA53 GA19 GA51 HA02 JA18

Claims (7)

[Claims] 1. An image pickup unit for picking up an image in a room, a video image picked up by the image pickup unit is divided into a plurality of feature amount calculation regions, and image feature amounts are picked up from the image picked up by the image pickup unit for each feature amount calculation region. Extraction means for extracting, detection means for detecting the temporal change amount of the image feature amount extracted for each feature amount calculation area at predetermined time intervals, temporal change amount of the image feature amount detected for each feature amount calculation area And a threshold for motion determination are compared for each feature amount calculation region,
If the temporal change amount of the image feature amount is less than or equal to the motion determination threshold value, it is determined that there is no motion in the feature amount calculation region, and if it is greater than the motion determination threshold value, there is motion in the feature amount calculation region. First motion determining means for determining, if the number of feature amount calculation areas determined to be in motion is less than or equal to a reference value, it is determined that there is no motion in the room, and the number of feature amount calculation areas determined to be in motion is a reference value If there is more motion, it is determined that the monitoring target has entered the room based on the second motion determination unit that determines that there is motion in the room, the determination result by the first motion determination unit, and the determination result by the second motion determination unit. Entry detecting means for detecting, a means for setting an entry state flag when the entry object detects that the monitoring target has entered the room, the determination result by the first motion determining means, and the second motion determining means. Exit detection means for detecting that the monitoring target has exited the room based on the determination result, and means for resetting the entry state flag when the exit target detects that the monitoring target has left the room, and An abnormality detection device, comprising: abnormality detection means for detecting an abnormality of a monitoring target based on the determination result by the second movement determination means when the entry state flag is set. 2. The entry detecting means determines that there is movement in the room by the second movement determining means and the first movement determining means determines that the entire image feature amount is within the calculation area when the entry state flag is reset. After it is determined that there is movement in the entry determination region set in the peripheral portion of the room, the second movement determination means causes the room to move after the first predetermined time or more has elapsed and before the second predetermined time has elapsed. When it is determined that there is movement within the room, and when it is determined by the first movement determination means that there is movement in the image feature amount calculation area in the central portion within the entire image feature amount calculation area, the monitoring target has entered the room. The abnormality detection device according to claim 1, wherein the abnormality detection device determines. 3. The entry determination area is fixed.
The abnormality detection device described in. 4. The abnormality detection device according to claim 2, wherein the entry determination area is determined based on a history of motion determination results by the first motion determination means at the time of previous entry detection. 5. The exit detection means determines that there is motion in the room by the second motion determination means and the first motion determination means within the entire image feature amount calculation area when the entry state flag is set. After a third predetermined time has elapsed since it was determined that there was movement in the central part of the above, the second movement determination means determined that there was movement in the room, and the first movement determination means determined that there was movement in the entry determination area. It is determined that the monitoring target exits from the room when it is determined that there is no movement in the room by the second movement determination means until the fourth predetermined time elapses. Item 5. The abnormality detection device according to any one of items 1, 2, 3 and 4. 6. The abnormality detection means determines whether or not an abnormality has occurred in the monitoring target in the room based on the past predetermined number of movement determination results by the second movement determination means when the entry state flag is set. The abnormality detection device according to any one of claims 1, 2, 3, 4 and 5, which determines whether or not. 7. The motion determination threshold value is set to a first threshold value when the entry state flag is reset, and is set to a second threshold value smaller than the first threshold value when the entry state flag is set. Claims 1, 2, 3, 4, which are set to a threshold value
The abnormality detection device according to claim 1, wherein the abnormality detection device according to claim 5.