JP4155159B2 - Combined human touch switch - Google Patents

Description

  The present invention relates to a composite human touch switch that automatically controls on / off of a load such as a lighting load in accordance with the presence / absence of a person in a room.

  Conventionally, a human touch switch that automatically turns on and off a load such as a lighting load according to the presence / absence of a person in a room such as a toilet or a conference room has been provided. As this type of human sensor switch, the heat rays radiated from the human body are monitored by a pyroelectric infrared sensor (hereinafter referred to as “pyroelectric element”). A passive type that detects absence is widely used. In other words, passive human sensors using pyroelectric elements do not need to project infrared rays, so that the circuit configuration is simple, can be provided at a small size and at a low price, and the field of view of a single unit is relatively Widely, even when a plurality of pyroelectric elements are arranged in parallel to expand the field of view, there is an advantage that they do not interfere with each other.

  However, since the pyroelectric element is a differential sensor that can obtain a voltage output corresponding to the rate of change (time change) of the received heat dose, when the pyroelectric element is used alone, the A voltage output is obtained only at the time of entry or exit, and it cannot be determined whether the entry or exit is within the field of view. Therefore, it is configured such that heat rays emitted from the human body enter the pyroelectric element via an optical element such as a light receiving lens or a light receiving mirror, and using the optical element causes uneven sensitivity within the field of view of the pyroelectric element. It is configured to occur. That is, as shown in FIG. 10A, a plurality of rectangular high-sensitivity regions DH that have higher sensitivity than the other regions are provided on the floor surface within the field of view of the pyroelectric element 10 provided on the ceiling. DH is arranged on a straight line in the width direction, or, as shown in FIG. 10B, a square-shaped high height that is more sensitive than the other areas on the floor surface within the field of view of the pyroelectric element 10 provided on the ceiling. A plurality of sensitivity areas DH are provided, and four high sensitivity areas DH are arranged in groups, and each group is arranged on a straight line, thereby giving uneven sensitivity in the field of view of the pyroelectric element 10. . In such a configuration, since a beam-like region connecting the pyroelectric element 10 and the high-sensitivity region DH on the floor surface has high sensitivity, the high-sensitivity beam-like region is called a detection beam BD.

  As described above, if a plurality of detection beams BD are formed in the field of view of the pyroelectric element 10, uneven sensitivity is formed in the field of view, so that a person existing in the field of view of the pyroelectric element 10 moves finely enough to move his / her hand. As a result, the heat dose incident on the pyroelectric element 10 changes with time, so that voltage output can be intermittently obtained from the pyroelectric element 10 during a period in which a person is present in the field of view of the pyroelectric element 10. become. Therefore, the configuration shown in FIG. 11 is widely adopted as an automatic switch that detects the presence or absence of a person using the pyroelectric element 10 and automatically turns on and off the load 16 according to the presence or absence of the person. In the configuration shown in FIG. 11, after the output of the pyroelectric element 10 is amplified by the amplifier 11, the comparator 12 removes noise below the reference voltage from the output of the amplifier 11 and shapes the waveform, thereby shaping the pyroelectric element 10. A rectangular wave signal is generated at the timing at which a voltage output is generated from, and this signal is used as a trigger for the timer circuit 14. The timer circuit 14 limits a delay time which is a fixed time when a trigger is given, and keeps the load 16 on through the load control circuit 15 during the delay time. The timer circuit 14 is retriggerable, and is configured to further limit the delay time from that point when a trigger is given from the comparator 12 during the delay time. Accordingly, when the person is detected by the pyroelectric element 10 and the load 16 is turned on and the time limit of the delay time is started, the load is detected every time the person is detected by the pyroelectric element 10 during the time limit of the delay time. The time for which 16 is continuously turned on is extended. As a result, assuming that the field of view of the pyroelectric element 10 is set indoors, the load 16 is on as long as there is a person in the room and human tremor is detected before the delay time expires. It is kept in. If no person is detected before the delay time expires due to the person leaving the room or the like, the timer circuit 14 expires and the load 16 is turned off.

  Assume that the pyroelectric element 10 detects whether a person is present in the room and turns on the load 16 such as a lighting fixture while the person is present. FIG. 12A shows the presence or absence of a person in the room. When a person is present in the room and moves or slightly moves within the field of view of the pyroelectric element 10, a voltage output is generated from the pyroelectric element 10, and FIG. As shown in b), the output of the amplifier 11 varies. As shown in FIG. 12C, the comparator 12 outputs a rectangular wave signal by removing noise from the output of the amplifier 11 and performing waveform shaping. The timer circuit 14 sets the delay time Td using the signal output from the comparator 12 as a trigger. Here, if a person is present in the field of view of the pyroelectric element 10 and is stationary without fine movement, the time limit of the delay time Td by the timer circuit 14 expires, as shown in FIG. The load 16 is turned off. That is, if the person's still state continues beyond the delay time Td, the load 16 is turned off despite the presence of the person in the field of view of the pyroelectric element 10.

  For example, when a luminaire is used as the load 16 in the toilet, the luminaire is turned on when a person enters the toilet, but when the person is stationary in the toilet reaches the delay time Td, Even if it exists, it causes a disadvantage that the luminaire is turned off. In order to avoid such inconvenience, when a lighting fixture is used as the load 16, as shown in FIG. 13 (d), step dimming is performed before the delay time Td expires, or FIG. For example, a technique for notifying that the expiration of the delay time Td is approaching by generating a notification sound by a buzzer or the like before the expiration of the delay time Td is known. If notification is made before the expiration of the delay time Td as described above, the person in the room can re-trigger the timer circuit 14 by moving, and the load 16 can be kept on.

  However, even if a technique for informing before the expiration of the delay time Td is adopted, a person who does not recognize the function of this type of human sensor cannot be caused to act so as to continue the load 16 being turned on. Therefore, it is difficult to adopt the above-described technology for public use because it causes anxiety due to dimming or notification sound. Therefore, the current situation is that the delay time Td is set relatively long (10 to 15 minutes) instead of providing the notification function.

  By the way, in the toilet for the disabled as a public facility, it is not necessary to turn on / off the lighting fixtures in the room, improving usability and realizing energy saving. In many cases, the lighting fixture is turned on, and the lighting fixture is turned off when not in use. However, if the delay time Td becomes longer as described above, the lighting fixture will continue to be lit for a relatively long time after the person leaves the toilet, resulting in unnecessary power consumption and energy saving. In addition, the lighting fixtures are lit even though the person has left the room, so that the next person who tries to use the toilet mistakes that the toilet is in use and keeps the lighting fixture off until it turns off. The inconvenience of waiting for a long time may occur.

  On the other hand, while using two pyroelectric elements, the field of the other pyroelectric element is set in an annular shape outside the field of one of the pyroelectric elements, and according to the temporal relationship of human detection by both pyroelectric elements, A technique for discriminating intrusion, residence, and exit has been proposed (see, for example, Patent Document 1). In this configuration, for example, after a person is detected by a pyroelectric element having an outer visual field as shown in FIG. 14 (a), the person is detected by a pyroelectric element having an inner visual field as shown in FIG. 14 (b). Then, as shown in FIG. 14 (c), it is determined that the room is entered, and then a person is detected by the pyroelectric element having the inner visual field, and then the person is detected by the pyroelectric element having the outer visual field. Then, it can be determined that the user leaves the room as shown in FIG.

A configuration in which an ultrasonic sensor having a detection region within the field of view of the pyroelectric element is also conceivable (see, for example, Patent Document 2). Ultrasonic sensors are devices that determine the presence or absence of an object from a change in distance by transmitting ultrasonic waves intermittently and receiving reflected waves and replacing the time from transmission to reception with distance. Use. When this configuration is adopted, after a person is detected by a pyroelectric element as shown in FIG. 15A, a person is detected by an ultrasonic sensor as shown in FIG. 15B, as shown in FIG. 15C. If it is determined that the person has entered the room and then a person is detected by the pyroelectric element after the person is no longer detected by the ultrasonic sensor, it can be determined that the person has left the room as shown in FIG.
JP 2001-291176 A JP-A-10-153656

  As described above, when a pyroelectric element is used alone, it is not possible to identify whether it is entering or leaving the field of view, so it is possible to reliably prevent lights from entering the room when using toilet lighting equipment as a load. If it tries to do so, the delay time must be lengthened, and if the delay time is lengthened, the inconvenience that the lighting apparatus continues to be lit for a relatively long time after leaving the toilet occurs.

  On the other hand, in a configuration in which a plurality of sensors are combined, it is possible to distinguish between entry and exit, so it is possible to keep the lighting fixtures on during the period from entry to exit, and from leaving the room to turning off the illumination fixtures. This time can be set relatively short, and it is considered that the problem in the case of using a pyroelectric element alone can be solved.

  However, in order to determine the order of detection, the order of detection of humans by both sensors (relationship) is used, and in a configuration in which two pyroelectric elements are combined, It is necessary to set a blank area in which no person is detected. Therefore, when a person moves along a blank area, it may not be possible to determine whether to leave the room. In addition, since the two pyroelectric elements must be set to large and small, it can be applied only to a relatively large space, and an appropriate field of view cannot be set when the space in the toilet is small There is also sex.

  Also, in the configuration in which the pyroelectric element and the ultrasonic sensor are combined, since the detection area of the ultrasonic sensor is set within the field of view of the pyroelectric element, a blank area is not formed, but ultrasonic waves are used when the space in the toilet is narrow. Multiple reflections may occur and people may not be detected accurately.

  After all, there is a problem that it is difficult to accurately determine the entrance / exit from either the configuration using two pyroelectric elements or the configuration combining the pyroelectric elements with an ultrasonic sensor.

  The present invention has been made in view of the above-mentioned reasons, and its purpose is to accurately turn on / off the room, so that the load is kept on while in the room and the load is turned off in a short time after leaving the room. The object is to provide a composite human sensor that realizes energy saving.

According to the first aspect of the present invention, a hot-wire human sensor that monitors a heat ray installed in a room and radiates from a human body and detects the presence / absence of a human by a change in heat dose, and starts after detection of a human by the hot-wire human sensor An image-type human sensor that picks up an indoor image and detects when a person in the image leaves the room, a switch circuit that turns on and off the load, and a human detection point and an image human The time point at which the sensor detects the exit of the room is used as a starting point for starting the delay time, and the switch circuit is kept on during the delay time, and the hot-wire sensor and the image sensor are detected within the delay time. if neither detects a human and a detection processing unit for turning off the switch circuit in the expiration time of the delay time, the field of view hot-wire motion sensor monitors the heat rays and the image type human cell The field of view captured by the camera is set to be equal, and the detection processing unit sets the delay time that is timed from the time of human detection by the hot-wire human sensor to a constant initial delay time and is constant from the initial delay time. Set the start time after subtracting the allowance time, start the image type human sensor when the start time has elapsed from the time of human detection by the hot wire type human sensor, and further from the time of detection of leaving the room by the image type human sensor The delay time that is limited is the exit delay time that is shorter than the initial delay time. During the exit delay time, the sensitivity of at least one of the hot-wire sensor and the image sensor is higher than the other periods. It is characterized by that.

According to this configuration, the hot-wire-type human sensor and the image-type human sensor do not detect the entrance / exit from the prior detection relationship (detection order) of the two, but the hot-wire human sensor detects the time when the room enters from the entrance / exit. Is detected without delay, and the time of exit from the entrance / exit is detected without delay by the image-type human sensor, so that the role of each sensor is shared, so that entrance / exit can be accurately determined . Moreover, since the field of view of the heat ray type human sensor and the field of view of the image type human sensor are matched, it can be used even in a small indoor space, and in addition, an active type such as an ultrasonic sensor. Therefore, there is no possibility of misidentification due to multiple reflections. Even when the exit from the room cannot be detected, the load is turned off at the end of the delay time. Therefore, even if the detection of the exit from the room fails, the load is automatically turned off to realize energy saving.
Furthermore, since the on-state of the load can be ensured at least for the initial delay time from the detection of the entrance by the hot wire type human sensor , the load is turned off during the entrance by setting the initial delay time relatively long. It is possible to prevent this . In addition, since the image-type human sensor is started after the start-up time from the time of human detection by the hot-wire human sensor, the time obtained by subtracting the margin time during the initial delay time is used as the start-up time. Compared with the case where the image type human sensor is activated immediately after human detection, power consumption by the image human sensor can be reduced, resulting in energy saving. Furthermore, since there is a margin time from the start of the image sensor to the expiration of the initial delay time, the operation of the image sensor can be stabilized during the margin time, and the load is kept on. While leaving, it becomes possible to detect exit.

Also, since the exit delay time after leaving the room is shorter than the delay time until leaving the room, it is easier to keep the load on while entering the room by setting a longer delay time from entering the room to leaving the room. Therefore, it is possible to prevent wasteful consumption of electric power by setting the exit delay time short. As a result, not only will it save energy, but it will be mistaken for being in use by leaving the lighting fixtures off for a long time after leaving the room when the toilet lighting fixtures for the disabled are used as loads. Can be prevented.

In addition, since the sensitivity of detecting a person in the exit delay time, which is the delay time after being judged to leave by the image type human sensor, is increased, it is ensured that no person remains in the room during the exit delay time. Even if it is mistaken to leave the room, the load can be kept on as long as there is a person in the room.

According to a second aspect of the invention, in the first aspect of the invention, the image-type human sensor is installed so as to have a downward visual field including at least an entrance, and a circumscribed rectangle is set in an area in which an image can be considered to exist. When the coordinates of the representative points of the circumscribed rectangle in the image move from the center to the periphery of the visual field of the image type human sensor, and the circumscribed rectangle disappears from the field of view, it is determined that the person leaves the room. It is characterized by doing.

  According to this configuration, when the image type human sensor is arranged on the ceiling and the visual field is set downward so as to include the entrance / exit in the visual field, the exit from the room can be easily detected. That is, it is possible to provide an exit detection technique based on an image suitable for use in a relatively small room.

According to a third aspect of the present invention, in the first or second aspect of the present invention, the image-type human sensor is installed with a visual field facing obliquely downward toward the doorway, and has a circumscribed rectangle in a region where a person can be considered to exist in the image. And at least one of the vertical and horizontal dimensions of the circumscribed rectangle in the image is reduced over time, and when the circumscribed rectangle disappears from the field of view and the area of the circumscribed rectangle is less than the specified threshold The point in time is determined as the time when the person leaves the room.

  According to this configuration, it is possible to easily detect exit when the visual field is set obliquely downward. That is, it is possible to provide an exit detection technique using an image suitable for use in a relatively large room.

According to a fourth aspect of the present invention, in the first to third aspects of the present invention, the image-type human sensor sets a circumscribed rectangle in an area in which an image can be regarded as a person, and the detection processing unit moves from entering the room to leaving the room. The maximum number of circumscribed rectangles generated in the image during this period is regarded as the number of people entering the room, and the time when the number of people leaving the room coincides with the number of people entering the room is set as the time of detection of leaving by the image type human sensor.

  According to this configuration, when a plurality of people enter and leave the room, it becomes possible to maintain the load on until all of them leave the room, even though the load operated when entering the room remains in the room. It can be prevented from being turned off.

According to the configuration of the present invention, to detect without delay entry point from the entrance by a hot-wire motion sensor, since detecting without delay exit point from the doorway by the image type human sensor, the role of each sensor are shared This makes it possible to accurately determine entry / exit . Moreover, since the field of view of the heat ray type human sensor and the field of view of the image type human sensor are matched, it can be used even in a small indoor space, and in addition, an active type such as an ultrasonic sensor. do not use the sensor, an advantage that does not occur even mistaken due to multiple reflection there Ru.
Furthermore, since the on-state of the load can be ensured at least for the initial delay time from the detection of the entrance by the hot wire type human sensor , the load is turned off during the entrance by setting the initial delay time relatively long. It is possible to prevent this . In addition, since the image-type human sensor is started after the start-up time from the time of human detection by the hot-wire human sensor, the time obtained by subtracting the margin time during the initial delay time is used as the start-up time. Compared with the case where the image type human sensor is activated immediately after human detection, power consumption by the image human sensor can be reduced, resulting in energy saving. Furthermore, since there is a margin time from the start of the image sensor to the expiration of the initial delay time, the operation of the image sensor can be stabilized during the spare time, and the load remains on. While leaving, it becomes possible to detect exit.
Also, since the exit delay time after leaving the room is shorter than the delay time until leaving the room, it is easier to keep the load on while entering the room by setting a longer delay time from entering the room to leaving the room. Therefore, it is possible to prevent wasteful consumption of electric power by setting the exit delay time short. As a result, not only will it save energy, but it will be mistaken for being in use by leaving the lighting fixtures off for a long time after leaving the room when the toilet lighting fixtures for the disabled are used as loads. Can be prevented.
In addition, since the sensitivity to detect people during the exit delay time is increased, it can be guaranteed that no people remain in the room during the exit delay time, and even if the exit is mistaken, there is a person in the room. If so, the load can be kept on.

  As shown in FIG. 1, the present invention includes a hot-wire human sensor 1 and a TV camera using a pyroelectric infrared sensor (hereinafter abbreviated as “pyroelectric element”) 10 for detecting a heat ray radiated from a human body. The presence / absence of a person in a specific area such as a room is determined by using together with the image-type human sensor 2 using the imaging means 20 as described above, and the load is determined according to the presence / absence of the person in the specific area. It controls on / off. In this embodiment, the inside of a toilet is assumed as a specific area, and the illumination load which illuminates the inside of a toilet is assumed as a load. In this type of application, if the lighting load is turned on when entering the toilet, the lighting load will continue to be turned on as long as there are people in the toilet, and the lighting load will be turned on immediately after leaving the toilet. However, it is required to turn off the lighting load in a relatively short time after leaving the room. As described in the conventional configuration, the pyroelectric element 10 detects a human intrusion into the detection area set in the field of view with a high probability, but the stationary state of the person in the detection area and the exit of the person from the detection area Is difficult to distinguish. Therefore, in the present invention, in order to compensate for the disadvantages of this type of pyroelectric element 10, the imaging means 20 is used in combination with human detection, the pyroelectric element 10 is mainly used to detect entry into the toilet, and the user leaves the toilet. For the detection, a configuration using the imaging means 20 is mainly employed.

  A lighting fixture 4 as a load is disposed in the toilet, and the lighting fixture 4 includes a lighting load (a lamp or a combination of a lamp and a lighting device). The composite human sensitive switch of the present embodiment includes a switch circuit 32 that is inserted into a power supply path between a commercial power source (not shown) and the lighting fixture 4, and a switch element (relay or 3) provided in the switch circuit 32. The lighting fixture 4 is turned on and off by opening and closing the terminal bidirectional thyristor. That is, since the switch circuit 32 includes a switch element, it can be used in place of a normal mechanical switch such as a wall switch. In addition, since the switch circuit 32 is connected to a commercial power supply, a configuration in which the power of the internal circuit is supplied through the switch circuit 32 is adopted. Therefore, it is possible to turn on / off the power supply from the commercial power supply to the lighting fixture 4 and to supply power from the commercial power supply to the internal circuit only by connecting the commercial power supply to one set of power supply terminals.

  The switch element provided in the switch circuit 32 is opened and closed by the control unit 31. As described above, in the present embodiment, the pyroelectric element 10 and the imaging unit 20 are used together to control the on / off of the illumination load, and human detection by the hot-wire human sensor 1 using the pyroelectric element 10 is performed. In addition, a detection processing unit 30 is provided to determine whether a person enters or leaves the toilet by combining the human detection by the image type human sensor 2 using the image pickup unit 20. That is, the detection processing unit 30 determines whether a person enters or leaves the toilet and gives the determination result to the control unit 31. Further, the control unit 31 generates a control signal that defines the timing of lighting and extinguishing of the luminaire 4 and controls the switch circuit 32 by this control signal. Also, if the load is an illumination load and the inside of the toilet is bright, there is no need to turn on the illumination load even if there are people in the toilet. A brightness sensor 13 composed of CdS or a photodiode to be detected is provided, and the detection processing unit 30 determines whether or not to turn on the luminaire 4 according to the brightness in the toilet detected by the brightness sensor 13. It also has a function to judge.

  Human detection by the hot-wire human sensor 1 using the pyroelectric element 10 is a well-known technique. The amplifier 11 amplifies the output of the pyroelectric element 10 and the comparator 12 outputs a signal corresponding to the level of human movement. To extract. The pyroelectric element 10 is a differential type sensor that can obtain a voltage output corresponding to the rate of change of the received heat dose (infrared ray amount). Since the output from the element 10 is small, this type of noise component is removed by using the comparator 12. The comparator 12 also has a waveform shaping function, and the output of the comparator 12 corresponding to the movement of the person detected by the pyroelectric element 10 has a rectangular wave shape. Although not shown, it is desirable to arrange a light receiving lens with uneven sensitivity in the field of view of the pyroelectric element 10 in front of the light receiving surface of the pyroelectric element 10. Even when the person moves his / her hand within the visual field of the electric element 10, the heat dose incident on the pyroelectric element 10 changes, and an output corresponding to the movement of the person is obtained from the comparator 12. Such a light receiving lens is formed as an assembly of small lenses that are converging lenses, and each small lens has high sensitivity in the area near the optical axis of each small lens and low sensitivity near the boundary of each small lens. That is, a beam-shaped detection area that is narrower than the visual field of the pyroelectric element 10 is set in the vicinity of the optical axis. Such a detection area in the vicinity of the optical axis of each small lens is called a detection beam. The pyroelectric element 10 is disposed on the ceiling above the toilet in the toilet, and has a field of view including at least the toilet.

  Both the output of the comparator 12 and the output of the brightness sensor 13, which are the outputs of the heat ray human sensor 1 described above, are input to the detection processing unit 30. In the detection processing unit 30, the presence or absence of a person in the field of view of the pyroelectric element 10 is determined using a rectangular wave signal input from the comparator 12. The detection processing unit 30 determines that a person is present in the field of view of the pyroelectric element 10 when, for example, a specified number (for example, three) of rectangular wave signals is input within a specified time (person detection). Judged). However, as described in the conventional configuration, the pyroelectric element 10 is a differential sensor and can only be detected as a person when the person is moving. When the determination is made, a timer circuit (off-delay timer) built in the detection processing unit 30 is activated, and the switch circuit 32 is kept on through the control unit 31 while the timer circuit limits the delay time. The timer circuit is retriggerable, and if it is determined that a person is detected during the timed operation of the timer circuit, the delay time is further limited from that point. The delay time of the timer circuit is variable as will be described later, and is normally set to be relatively long so that the switch circuit 32 is kept on even when there is no movement of the person while the person is in the toilet. is there.

  A threshold for the output of the brightness sensor 13 is set in the detection processing unit 30, and a switch circuit 32 is provided to the control unit 31 during a period in which the brightness (illuminance) detected by the brightness sensor 13 exceeds the threshold. Instruct to keep off. Further, the detection processing unit 30 provides hysteresis to the brightness sensor 13 using a two-stage threshold value, and when the brightness detected by the brightness sensor 13 is equal to or lower than the lower threshold value, The switch circuit 32 corresponding to human detection is turned on / off until the higher threshold value is exceeded, and then the switch circuit 32 is turned off until the brightness falls below the next lower threshold value when the brightness exceeds the higher threshold value. To keep. Therefore, even if the brightness change vibrates near any threshold value, the operation is stabilized. Here, during the period when the switch circuit 32 is turned on / off, when the lighting fixture 4 is turned on in response to human detection, the brightness detected by the brightness sensor 13 may exceed the higher threshold. When the switch circuit 32 is turned on in response to the detection, the output of the brightness sensor 13 is disabled, thereby preventing a malfunction in which the lighting fixture 4 is repeatedly turned on and off in a short time.

  By the way, the image pickup means 20 is a solid-state image pickup device such as a CCD image sensor or a CMOS image sensor, and outputs images picked up at predetermined time intervals. The analog signal of the image output from the imaging unit 20 is converted into a digital signal by A / D conversion in the image input unit 21. However, when a CMOS image sensor having a function of outputting a digital signal is used for the imaging unit 20, A / D conversion in the image input unit 21 is not necessary. A color image can be used as an image picked up by the image pickup means 20, but a monochrome grayscale image is adopted here. The time interval captured by the imaging unit 20 may be set as appropriate within a range in which the presence or absence of a moving object can be determined from the time-series images obtained at the time interval, and is not intended to obtain a smooth moving image. Therefore, it is not necessary to output an image of 30 frames per second.

  The output of the image input means 21 is input to the contour extraction means 22, and the grayscale image which is the output of the image input means 21 is input to the contour extraction means 22, and the differential value and direction code of each pixel are obtained from the grayscale image. That is, the contour line extraction unit 22 obtains a differential image in which the pixel value of each pixel is a differential value and a direction code image in which the pixel value of each pixel is a direction code, and the differential image and the direction code image are shaded. Along with the image, it is stored in a storage unit 23 comprising a semiconductor memory.

  Various methods for obtaining the differential value have been proposed. Basically, for the neighboring pixels of the pixel of interest (8 neighborhoods are widely adopted), the luminance difference in the vertical direction of the image is the luminance difference in the horizontal direction. The divided value is used as a differential value. However, the differential image is generated from the grayscale image by using the fact that the differential value becomes large near the boundary between the object and the background due to the difference in the brightness value between the object and the background in the image, and the object outline candidate In this embodiment, weighted differentiation using a Sobel filter is performed to enhance the contour line.

  The direction code is a value in which the differential value is associated with the change direction of the luminance value, and an integer value code is associated in 8 directions in units of 45 degrees (in this case, from the neighboring pixels of 8). The direction code is associated with the obtained normal differential value). The direction code of each pixel is set so as to represent a direction orthogonal to the direction in which the change of the luminance value is maximized in the image. Therefore, the direction indicated by the direction code in each pixel substantially coincides with the extending direction of the contour line (the three pixels adjacent to each other within the range of ± 45 degrees with respect to the direction indicated by the direction code of each pixel are pixels on the contour line of the object. Is likely to be).

  In the differential image obtained by the contour extracting means 22 as described above, a portion having a large contrast is emphasized. Therefore, by binarizing the differential image with an appropriate threshold, candidates for the contour line of the object included in the differential image are obtained. Can be extracted. The contour extracting unit 22 thins the extracted region that is a candidate for a contour line to a width of one pixel, and extracts a candidate for an edge that is a candidate for a contour line. Since the edge candidates may be interrupted, the pixels are tracked using the direction code for the edge candidates, and a contour image composed of the edges obtained by connecting the edge candidates that can be regarded as the contour lines of the object is generated and stored. Stored in the contour image storage means 23 a provided in the unit 23 for a certain period. The storage unit 23 is also used as a work area when obtaining a contour image.

  The contour image stored in the contour image storage unit 23a is input to the moving contour extracting unit 24, and the moving contour extracting unit 24 extracts edges corresponding to the moving object from three or five contour images. Here, a technique for extracting an edge corresponding to a moving object from three contour images will be described with reference to FIG. Now, as shown in FIGS. 2A to 2C, three contour images E (T−ΔT), E (T), E (T + ΔT) captured at times T−ΔT, T, and T + ΔT are obtained. It is given to the moving contour extraction means 24. In the illustrated example, contour images E (T−ΔT), E (T), and E (T + ΔT) each including the moving object Ob are shown.

  The moving contour extracting means 24 first obtains a difference between each pair of contour images adjacent in time series (that is, E (T−ΔT) and E (T), E (T) and E (T + ΔT)) ( Since this image is a difference between the contour images, it is hereinafter referred to as a “difference contour image”). However, since the contour image is a binary image having different pixel values in the edge portion and the portion other than the edge, the moving contour extraction unit 24 performs the pair of contour images for each pair of pixels at the same position. If the logical operation for obtaining the exclusive OR is performed, the difference between the pair of contour images of interest is obtained. In the two differential contour images obtained from the contour image in the illustrated example, the moving object Ob appears twice in each differential contour image.

  In the moving contour extracting means 24, in order to extract the moving object Ob included in the contour image E (T) at time T, a logic for obtaining a logical product for each pair of pixels at the same position for the two differential contour images. The calculation is performed, and the resulting image is output as a candidate image as shown in FIG. That is, since the background is almost removed from the two differential contour images, the background of the contour image E (T) at time T, which is a common part, is obtained by performing a logical product operation on the two differential contour images. It is considered that a candidate image obtained by removing is obtained, and this candidate image only includes noise in addition to the moving object Ob.

  Here, in this embodiment, an example using three contour images E (T−ΔT), E (T), and E (T + ΔT) is shown, but candidate images using four or more contour images. Can also be generated. For example, when five contour images E (T-2ΔT), E (T−ΔT), E (T), E (T + ΔT), and E (T + 2ΔT) are used, first, two contour images are used. For (E (T−2ΔT) and E (T + 2ΔT), E (T−ΔT) and E (T + ΔT)), a contour image of the background from which the moving object Ob is removed by a logical operation for obtaining a logical product is generated. When the logical operation for obtaining the logical product of the two contour images obtained in this manner and inverting the contour image E (T) is performed, the contour images E (T−2ΔT) and E (T + 2ΔT) are obtained. The contour image including the moving object Ob in the background and the contour image E (T), and the contour image E (T−ΔT) and E (T + ΔT) are hidden by the moving object Ob. A background and a contour image including the moving object Ob in the contour image E (T) are obtained. If a common part is extracted by a logical operation for obtaining a logical product for both contour images, a contour image (candidate image) including the edge of the moving object Ob in the contour image E (T) is obtained. In addition, candidate images can be generated by variously combining four or more contour line images.

  In the candidate image, a logical operation is performed on a binary contour image instead of obtaining a difference from the grayscale image, and three or more contour images are extracted instead of extracting the moving object Ob from the two images. Since the moving object Ob included in the contour image at a specific time is extracted using the same moving object Ob does not appear in two places in the candidate image, a change including the moving object Ob occurred. Only the region can be extracted. Since the candidate image output from the moving contour extracting unit 24 includes noise in addition to the moving object Ob, the candidate image obtained by the moving contour extracting unit 24 is input to the moving region detecting unit 25, and the moving region detecting unit 25. Then, labeling is performed for each area where pixels are connected (connected area). Here, if a circumscribed rectangle D1 is set for each connected region as shown in FIG. 2 (e) and labeling is performed on the circumscribed rectangle D1, compared with the case where a label is assigned to each pixel. The amount of data can be reduced. In addition, since the candidate image is obtained using three or more outline images as described above, there is no afterimage as in the case of obtaining the difference between frames in the grayscale image, and only the edge of the moving object is extracted. It becomes possible to do.

  The candidate image as shown in FIG. 2D output from the moving contour extraction unit 24 and the direction code image stored in the storage unit 23 are input to the region feature amount detection unit 26. Then, it is evaluated whether the area given the label by the moving area detecting means 25 is an area corresponding to a person or a disturbance other than a person. In the area feature quantity detection means 26, first, for each area labeled in the candidate image obtained as the output of the moving contour extraction means 24, the edge is referenced with reference to the direction code image stored in the storage unit 23. The direction code of the upper pixel is obtained, and the frequency distribution related to the direction code is obtained for each labeled region. The frequency distribution is normalized by the total number of pixels on each target edge. In addition, the direction code does not use eight types of direction codes, but direction codes that are opposite to each other in the same direction are regarded as the same direction code, and a frequency distribution is obtained for the four types of direction codes. That is, a direction code corresponding to 0 degrees and 180 degrees, a direction code corresponding to 45 degrees and 225 degrees, a direction code corresponding to 90 degrees and 270 degrees, a direction code corresponding to 135 degrees and 315 degrees, and The four types of direction codes are used. 3, for one region, 0 degree and 180 degrees are “direction 1”, 45 degrees and 225 degrees are “direction 2”, 90 degrees and 270 degrees are “direction 3”, 135 degrees and 315 degrees. An example of the frequency distribution obtained by classifying as “direction 4” is shown as a histogram.

  The area feature quantity detection means 26 obtains a frequency distribution for each labeled area, and then determines whether it is a person or a disturbance based on the shape of the frequency distribution. In this judgment, the edge corresponding to the person has more curved parts than the straight line part, and the edge corresponding to the person has a complicated shape, so the appearance frequency is compared for all direction codes in the pixels above the edge. On the other hand, an empirical rule is used that an edge due to a structure or the like has a lot of straight line portions and often shows a distribution biased in a specific direction. In other words, regarding the frequency of each direction code when the region corresponds to a person, a normal range is set for each direction code with an upper limit value and a lower limit value, and the frequency distribution obtained for each region is out of the frequency of each direction code. If any one of them deviates from the normal range, the region is regarded as a disturbance other than a person.

  Further, the area feature quantity detection means 26 evaluates whether or not the area includes a person for an area determined not to have an extreme distribution bias in the frequency distribution (an area that has not been regarded as a disturbance). In other words, using the reference data in which the frequency distribution of the edge direction code related to the person is registered in advance as a table in the storage unit 23, the frequency distribution for each region that is not regarded as a disturbance is compared with the frequency distribution of the reference data. , Evaluate the similarity between the two. Although the similarity between two frequency distributions can be evaluated by various methods, the similarity can be evaluated with a simple method by using the square sum of the frequency differences in each direction as an evaluation value. This evaluation value is compared with an appropriately set threshold value, and if the evaluation value is equal to or less than the threshold value, the area is determined as an area corresponding to a person. In this method, the amount of reference data is small and the amount of comparison is small compared to the case of template matching.

  As described above, a region corresponding to a moving object can be extracted from the image captured by the image capturing unit 20, and the region feature amount detection unit 26 determines whether the region corresponding to the moving object is a region corresponding to a person. Since the determination can be made, it is possible to detect the presence or absence of a person based on the image captured by the imaging unit 20. That is, the image-type human sensor 2 that detects a person based on the image captured by the imaging unit 20 is configured by a series of processes from the imaging unit 20 to the region feature amount detection unit 26. The determination result by the image type human sensor 2, that is, the determination result by the region feature amount detection means 26 is input to the detection processing unit 30. The moving contour extracting unit 24, the moving region detecting unit 25, the region feature amount detecting unit 26, the detection processing unit 30, and the control unit 31 are realized by using a microcomputer and constitute the signal processing unit 3 together with peripheral circuits. Yes.

  Next, in the detection processing unit 30, a procedure for controlling the switch circuit 32 using human detection by the hot-wire human sensor 1 and human detection by the image human sensor 2 (that is, the imaging unit 20) will be described. Hereinafter, as illustrated in FIG. 4A, a description is given assuming a period in which the brightness (illuminance) detected by the brightness sensor 13 does not exceed the higher threshold after being lower than the lower threshold. To do. As described above, since this period is a period during which the detection processing unit 30 can control the on / off of the switch circuit 32, the pyroelectric element 10 receives the heat rays emitted from the human body and changes to the received heat dose. As a result of this, a rectangular wave output as shown in FIG. 4B is obtained from the comparator 12, and as shown in FIG. 4C, the detection processing unit 30 determines human detection.

  In the example shown in FIG. 4, as shown in FIGS. 4C and 4D, the activation signal Sa is output from the detection processing unit 30 when it is determined that the person is detected based on the output of the heat ray human sensor 1. The image type human sensor 2 is activated. When the image type human sensor 2 is activated, as shown in FIG. 4E, the image type human sensor 2 determines that the person is detected. However, the image type human sensor 2 detects a moving object as described above, and if the human motion stops in the toilet, there is a possibility that the image type human sensor 2 may not determine human detection. (In the illustrated example, the image-type human sensor 2 represents a state in which human detection is continuously determined), and the hot-wire human sensor 10 may not be determined as human detection. Therefore, by setting the delay time of the timer circuit to be relatively long as described above, the lighting fixture 4 is prevented from being turned off while a person is present in the toilet.

  On the other hand, when a person leaves the toilet, movement occurs in the person. Therefore, both the hot-wire human sensor 1 and the image human sensor 2 are determined to be human detection. However, since it is difficult to accurately detect the time when the person leaves the toilet in the determination of human detection by the hot-wire human sensor 1, as shown in FIG. Judgment is made by the sensor 2. A method of determining leaving by the image type human sensor 2 will be described later. When a person leaving the toilet is detected by the image-type human sensor 2, a delay time Td2 is set in which the delay time of the timer circuit is shorter than the initial delay time Td1 during the period in which the person stays in the toilet. Then, as shown in FIG. 4F, when the exit delay time Td2 elapses from the exit, the switch circuit 32 is turned off (the lighting fixture 4 is turned off). The leaving delay time Td2 after leaving the room is set to the shortest delay time (about 10 seconds) that can be set in the detection processing unit 30. Since the timer circuit is retriggerable, even when the initial delay time Td1 is in the time limit, if the image-type human sensor 2 detects the exit, the delay time is changed to the exit delay time Td2 and the timed operation is performed. The image type human sensor 2 continues to operate until the expiration time of the exit delay time Td2.

  By the way, the determination of the person detection by the heat ray type human sensor 1 is for the purpose of detecting the time of entering the toilet, while the determination of the human detection by the image type human sensor 2 is for the purpose of detecting the time of leaving the toilet. Therefore, it can be said that the image-type human sensor 2 may be activated before the person leaves the toilet. Therefore, from the point of determination of human detection by the hot-wire human sensor 1 as shown in FIG. 5C, the activation signal Sa is sent from the detection processing unit 30 after a certain activation time Ta has elapsed as shown in FIG. 5D. It is desirable to output (see FIG. 1) and activate the image type human sensor 2. The activation time Ta is set to be shorter than the time when it is assumed that a person leaves the toilet from the time of determination of human detection by the hot-wire human sensor 1.

  In this way, if the activation of the image type human sensor 2 is delayed by the activation time Ta with respect to the time point when the human detection by the heat ray type human sensor 1 is determined, the person of the image type is detected at the time point when the human detection by the heat ray type human sensor 1 is determined. The power supplied to the image type human sensor 2 can be reduced as compared with the case where the sensor 2 is activated. The activation time Ta is desirably set based on a delay time set immediately after the determination of human detection by the hot-wire human sensor 1. In other words, at the time when the lighting fixture 4 is turned on by the human detection determination by the hot-wire human sensor 1, a relatively long initial delay time Td1 corresponding to the field condition is set in the detection processing unit 30, and this initial delay is set. It is desirable to use a value obtained by subtracting a certain margin time Tb from the time Td1 as the activation time Ta.

  When the activation time Ta is set in this manner, the activation time Ta is set based on the initial delay time Td1 according to the field conditions, and the image is previously displayed by a certain margin time Tb with respect to the initial delay time Td1. Since the expression type human sensor 2 is activated, it is possible to detect the exit from the toilet while the operation of the image type human sensor 2 is stable. Furthermore, the lighting apparatus 4 is not turned off unless the human detection is detected by the image type human sensor 2 during the initial delay time Td1 from the determination of human detection by the heat ray human sensor 1, and the lighting apparatus 4 is not turned off. The image-type human sensor 2 can be activated before the light is turned off. In short, after the determination of human detection by the hot-wire-type human sensor 1 and before the image-type human sensor 2 is activated (during the activation time Ta), the person leaves the toilet. Even when a person leaving the room cannot be detected, the lighting apparatus 4 can be turned off and the image-type human sensor 2 can be stopped after the initial delay time Td1 expires. In addition, by appropriately setting the margin time Tb, it is possible to secure the time from the activation of the image type human sensor 2 until the determination of the human detection can be surely performed.

  Here, assuming that there is no margin time Tb, since there is no person at the time when the human detection by the image type human sensor 2 can be determined, the image type human sensor 2 detects the exit of the room. In addition, since the initial delay time Td1 has also expired, the lighting fixture 4 cannot be turned off after the lighting fixture 4 is turned on by the determination of human detection by the hot-wire human sensor 1, and the lighting fixture 4 While the lighting state is continued, the operation of the image type human sensor 2 is also continued, and power is wasted. Therefore, as described above, it can be said that it is desirable to set the relationship of (start-up time Ta) = (initial delay time Td1) − (room time Tb).

  It is desirable that the field of view of the pyroelectric element 10 and the field of view of the imaging means 20 include the entire toilet. If the field of view of the imaging means 20 includes the entire interior of the room, the extinction time of the circumscribed rectangle D1 can be determined as the exit time, and even if a person moves to a place other than the entrance / exit, it is erroneously determined as the exit Because there is nothing to do. Further, it is desirable that the boundary of the visual field of the pyroelectric element 10 is made coincident with the entrance / exit of the toilet, so that when a person enters the visual field of the pyroelectric element 10, that is, when a person enters the toilet, the illumination is performed. It becomes possible to light the appliance 4 without delay. On the other hand, it is desirable that the boundary of the field of view of the image pickup means 20 coincides with the entrance / exit of the toilet or be extended slightly outside the entrance / exit of the toilet. With this setting, the heat-type human sensor 1 does not make a human detection determination after the image-type human sensor 2 detects the exit of the room, and the delay time can be ensured by the detection of the time when the image-type human sensor 2 leaves the room. Can be shortened.

  By the way, even if the exit time is detected by the image-type human sensor 2 as described above, there is still a possibility of false detection. Therefore, in the exit delay time Td2 after the detection of the exit time by the image-type human sensor 2, the toilet is It is necessary to continue to monitor the existence of people inside. That is, as described above, the image type human sensor 2 continues to operate until the exit delay time Td2 expires. Further, as long as the brightness detected by the brightness sensor 13 satisfies the condition (from the lower threshold value or less until the higher threshold value is exceeded), the monitoring by the hot-wire human sensor 1 is continued. Therefore, the person in the toilet is detected even in the leaving delay time Td2. Therefore, the presence / absence of a person is monitored even in the exit delay time Td2, but if no person is detected in the exit delay time Td2, the lighting apparatus 4 is turned off after the exit delay time Td2 expires, so the exit delay time Td2 It is necessary to ensure that no one is present in the toilet. Therefore, as shown in FIG. 5G, in the exit delay time Td2, the sensitivity of at least one of the heat ray type human sensor 1 and the image type human sensor 2 is increased more than usual so that the exit delay time Td2 It is desirable to determine that no one is in the toilet.

  The sensitivity of the hot-wire human sensor 1 can be adjusted by the gain of the amplifier 11 and the threshold value of the comparator 12. That is, in the exit delay time Td2, the sensitivity is increased by performing at least one of increasing the amplification factor of the amplifier 11 more than usual and lowering the threshold value of the comparator 12 more than usual. In addition, the sensitivity of the image type human sensor 2 can be adjusted by a threshold value to be compared with an evaluation value used for similarity evaluation in the region feature amount detection unit 26, and in the region labeled by the moving region detection unit 25. It can also be adjusted by increasing / decreasing the number of pixels used for evaluation of similarity. In other words, if the threshold value to be compared with the evaluation value is reduced, those with low similarity are regarded as similar, and if the number of pixels used for evaluation of similarity is reduced, the similarity is high and low And the difference in evaluation value becomes smaller. As a result, the error rate of human detection judgment is increased, but the leakage rate is reduced. This corresponds to substantially increasing the sensitivity of the image type human sensor 2.

  As described above, in the exit delay time Td2, the sensitivity of at least one of the hot-wire human sensor 1 and the image human sensor 2 is made higher than usual to ensure the presence or absence of a person in the exit delay time Td2. Even if there is a person in the toilet and the image-type human sensor 2 mistakenly detects that the person has left the room, it is possible to avoid turning off the lighting device 4. When it is determined that the person is detected in the exit delay time Td2, the operation returns to the operation when the human detection sensor 1 first determines that the person is detected.

  One of the following two types of methods is adopted for the determination of leaving the room by the image type human sensor 2 described above. Here, it is assumed that the imaging means 20 is installed near the center of the ceiling of the toilet, and the visual field VF is set as shown in FIG. The visual field VF includes a toilet 6 and a doorway (not shown, but a toilet doorway is present in the lower right of FIG. 6). Since the person moves when leaving the toilet, it is judged whether or not the person is leaving the room by monitoring the movement of the circumscribed rectangle D1 set by the moving area detecting means 25 as shown in FIG. That is, for the determination of leaving the room, the movement of the representative point (center point or vertex) of the circumscribed rectangle D1 is monitored, and the position of the representative point (circumscribed rectangle D1) is moved from the center of the field of view VF as shown in FIG. It is determined that the time when the person moves out of the toilet and the circumscribed rectangle D1 disappears from the field of view VF is when the person leaves the toilet. An arrow R1 in FIG. 6B indicates the direction in which the representative point has moved. Note that it is only necessary to detect entry into and exit from the toilet, and monitoring staying in the toilet is not essential. Therefore, when the toilet is large, the pyroelectric element 10 and the imaging means 20 are connected to the ceiling of the toilet. However, it may be arranged at a position biased toward the entrance / exit side of the toilet.

  The above-described configuration example assumes the case where the optical axis of the image type human sensor 2 is vertically downward, but the optical axis of the image type human sensor 2 may be inclined obliquely with respect to the vertically downward direction. For example, when monitoring the exit of a person from a conference room or the like that is larger than a toilet, the imaging means 20 is installed at a location away from the entrance and the optical axis is arranged toward the entrance. That is, since the optical axis is obliquely downward toward the entrance / exit, as shown in FIGS. 7A and 7B, as the person M approaches the entrance / exit EN, at least one of the vertical dimension and the left / right dimension of the circumscribed rectangle D1. One will shrink with time. More specifically, when the person M moves toward the entrance / exit EN, as shown in FIG. 7A, the lower side of the circumscribed rectangle D1 moves upward in the field of view VF and the area of the circumscribed rectangle D1 decreases. Further, when the person M moves along the wall surface provided with the entrance / exit EN, the left and right dimensions are reduced when leaving the room as shown in FIG. Therefore, the point in time when the circumscribed rectangle D1 disappears from the field of view VF and the point in time when the area of the circumscribed rectangle D1 becomes equal to or less than the specified threshold are determined as the time of leaving the room. That is, when the circumscribed rectangle D1 satisfies either of the two conditions in the image, it is determined that the user leaves the room. In this manner, the detection processing unit 30 can determine whether to leave the room by monitoring the circumscribed rectangle D1.

  Regarding the determination of leaving the room, the two methods described above can be applied when one person enters or leaves the room, but if there are multiple people in the room, the person remains in the room even after one person leaves the room. The lighting fixture 4 must be kept lit until the person leaves the room. Such a situation is not limited to a large room such as a conference room. For example, as shown in FIG. 8, only the caregiver M1 leaves the caregiver M1 after entering the toilet accompanying the care receiver M2. Also occurs. In such a situation, if a person who has left the room is still when the image-type human sensor 2 detects a person leaving the room, either the hot-wire human sensor 1 or the image human sensor 2 will be used. Since no person remaining in the room is detected, the lighting fixture 4 is turned off when the exit delay time Td2 elapses from the detection of the exit by the image type human sensor 2. That is, there is a possibility that inconvenience that the luminaire 4 is turned off in spite of a person remaining in the room.

  Therefore, in the period from when the person entering the toilet is detected by the heat ray type human sensor 1 until the image type human sensor 2 detects the exit, the movement area detecting means 25 in the image type human sensor 2 obtains it. The maximum number of circumscribed rectangles D1 in one screen is obtained, and each time a person leaves the room, the number of circumscribed rectangles D1 corresponding to the person leaving the room is subtracted from the maximum number of circumscribed rectangles D1, and this value finally becomes zero. At that time, it is considered that everyone leaves the room. In the example shown in FIG. 8, two circumscribed rectangles D1 are formed when the caregiver M1 moves away from the cared person M2 in the state of FIG. 8A, and one person (the caregiver M1 in FIG. 8B). Even if the cared person M2 is stationary and the circumscribed rectangle D1 is not formed (the circumscribed rectangle D1 is not formed is indicated by a broken line), one person is detected. Can be recognized in the toilet.

  Thus, by finding the maximum number of circumscribed rectangles D1, a value corresponding to the number of people entering the room is obtained, and the number of people leaving the room is subtracted. The possibility of turning off the lighting fixture 4 can be reduced. Here, since the maximum number of circumscribed rectangles D1 is obtained in a period from when a person enters the room to when the exit is detected, it is assumed that a circumscribed rectangle D1 in which a plurality of people overlap at the time of entering the room is obtained. In addition, it is considered that the number of circumscribed rectangles D1 corresponding to the number of people is obtained during the period, and the maximum number of circumscribed rectangles D1 is surely greater than or equal to the number of people entering the room. On the other hand, when only one circumscribed rectangle D1 is obtained by overlapping circumscribed rectangles D1 corresponding to a plurality of persons leaving the room, the number of circumscribed rectangles D1 corresponding to the left persons is subtracted from the maximum number of circumscribed rectangles D1. May not be zero. However, even in such a case, since the luminaire 4 is turned off after the delay time elapses, the purpose of continuing the lighting state of the luminaire 4 until everyone leaves the room can be achieved.

  As described above, as a technique for estimating the number of people entering the room, the maximum number of circumscribed rectangles D1 in the period from entering the room to leaving the room is regarded as the number of people entering the room, and the number of people entering the room is changed by changing the area of the circumscribed rectangle D1. It is also possible to estimate the number of people. For example, in the toilet for the handicapped, as shown in FIG. 9A, if the caregiver M1 and the care recipient M2 enter the room at the same time, the circumscribed rectangle D1 is 1 even though there are two people entering the room. Only one may be formed. In such a situation, as shown in FIG. 9 (b), the caregiver M1 is considered to leave the room with the cared person M2 left in the toilet, and the cared person M1 moves away from the cared person M2. If the movement of M2 is small, the circumscribed rectangle D1 may be formed only by the caregiver M1. That is, it is conceivable that only one circumscribed rectangle D1 can be obtained even though two people enter the room. However, since the imaging means 20 images the inside of the toilet from the ceiling side, the area of the circumscribed rectangle D1 formed by the movement of the caregiver M1 is smaller than the area of the circumscribed rectangle D1 when entering the room. Therefore, when the area of the circumscribed rectangle D1 changes by more than a prescribed threshold and the area is reduced, it is determined that one person leaves the room and one person remains. In addition, if a range for one person is defined for the area change of the circumscribed rectangle D1, it can be estimated that a plurality of persons overlap when the area deviates from the area occupied by one person. The area of D1 can be used as information for estimating the number of people.

  If the above-mentioned technology is adopted, it is possible to determine whether to leave the room, but if the room occupant is almost stationary or if the room occupant is hidden by a shield such as a pillar, As a result, no valid candidate image is obtained from the moving contour extracting unit 24, and the circumscribed rectangle D1 is not generated in the moving region detecting unit 25. Therefore, a circumscribing rectangle is set for the contour image obtained by the contour extraction means 22 after the entry by the heat ray human sensor 1 is detected, similarly to the moving area detection means 25, and the contour images at different times are circumscribed. The amount of movement of the rectangle is equal to or less than the specified threshold, and the number of pixels whose luminance value has changed among the pixels included in the circumscribed rectangle is less than the specified value, or the direction code of the pixel included in the circumscribed rectangle is obtained. When the distance of the frequency distribution (the evaluation value of the frequency distribution obtained by the same method as that of the region feature amount detection unit 26) is smaller than the specified value, the detection processing unit 30 considers the possibility of leaving the room, If it is determined that the user has left the room under the above-described conditions, it is determined to be stationary or shielded. In short, even if a valid candidate image cannot be obtained from the moving contour extraction means 24, it is not immediately determined to leave, but if it is not determined to leave under the above-mentioned conditions, it is stationary or shielded and stays in the room. (Corresponding to “residence” in the conventional example). When it is determined that the object is stationary or shielded, the image-type human sensor 2 continues to monitor the moving object, and the lighting device 4 of the lighting apparatus 4 is kept until the initial delay time set by the human-wire detection sensor 1 is determined. Keep on. In short, unless a clear exit determination is made, it is determined that it is not the exit, and the possibility that the lighting device 4 is turned off is reduced.

  By the way, in a room where disturbance light changes, there may be a region where the luminance changes relatively large in the field of view of the imaging means 20. When the indoor space is relatively small like a toilet, the area occupied by a person in the field of view of the imaging means 20 is large. Therefore, if a lower limit is provided for the area of the circumscribed rectangle D1, the area of noise due to ambient light is increased. The part is considered removable. However, when the indoor space is relatively wide like a conference room, the range of the circumscribed rectangle D1 cannot be narrower than the person existing far away from the imaging means 20, and therefore noise removal by the area of the circumscribed rectangle D1 is eliminated. Have difficulty.

  As a technique for solving this type of problem, it is conceivable to perform pattern matching with a template using an outline image, but in order to perform pattern matching, a large amount of memory is required for work. There arises a problem that a high-cost apparatus with high processing capability is required.

  By the way, as described above, in order to detect a leaving room using an image, since the moving direction and dimensional change of the circumscribed rectangle D1 with the passage of time are used, it is necessary to store a plurality of images. Even when the person is stationary, if noise is included, an area where it is determined that the person moves is generated, so that a plurality of images are accumulated and stored. However, since it is useless to accumulate images due to noise, it is necessary to discard the images from the storage unit 23 every certain discard time after the accumulation of images is started. When the image is discarded from the storage unit 23 in this manner, the image including the person is also discarded, and it is impossible to determine whether the person is moving or stationary immediately after the discarding. In this case, it is assumed that the person is stationary, and as described above, it is determined that the person is stationary until the exit is clearly detected. By determining in this way, it is possible to reduce the possibility of determining that the user has left the room even if the user has not left the room, and to reduce the possibility that the luminaire 4 is inadvertently turned off while in the room. it can.

  By the way, since the pyroelectric element 10 and the imaging means 20 are housed in the same case, when the optical axis of the imaging means 20 is set obliquely downward in a relatively large room such as a conference room, The pyroelectric element 10 moves away from the entrance and exit, and it may be difficult for the pyroelectric element 10 housed in the same case as the imaging unit 20 to obtain sufficient sensitivity to changes in heat dose from the vicinity of the entrance and exit. Therefore, in the present embodiment, as shown in FIG. 1, it is possible to add a handset 5 including a handset case that is independent from the case housing the pyroelectric element 10. The subunit | mobile_unit 5 is equipped with the structure corresponded to the pyroelectric element 10, the amplifier 11, and the comparator 12, and the output of the subunit | mobile_unit 5 becomes a rectangular wave shape corresponding to a human motion similarly to the output of the comparator 12. FIG. The output of the slave unit 5 is input to the detection processing unit 30, and the detection processing unit 30 determines whether to detect the person using the logical sum of the output of the comparator 12 and the output of the slave unit 5. Therefore, even if the case containing the image-type human sensor 2 and the pyroelectric element 10 is disposed at a location far from the entrance / exit, the handset 5 can be disposed near the entrance / exit, and detection of a person near the entrance / exit Can be performed by the handset 5. That is, the field of view of the pyroelectric element 10 is enlarged. The subunit | mobile_unit 5 can use not only one unit but multiple units | sets in parallel, and can attach the number of subunit | mobile_units 5 suitably according to a room shape.

It is a block diagram which shows embodiment of this invention. It is operation | movement explanatory drawing of the image type human sensitive sensor in the same as the above. It is operation | movement explanatory drawing of the image type human sensitive sensor in the same as the above. It is operation | movement explanatory drawing same as the above. It is operation | movement explanatory drawing same as the above. It is operation | movement explanatory drawing same as the above. It is operation | movement explanatory drawing same as the above. It is operation | movement explanatory drawing same as the above. It is operation | movement explanatory drawing same as the above. It is a figure which shows the example of a setting of the detection beam in a prior art example. It is a block diagram same as the above. It is operation | movement explanatory drawing same as the above. It is operation | movement explanatory drawing same as the above. It is operation | movement explanatory drawing of another prior art example. It is operation | movement explanatory drawing of another prior art example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Heat ray type human sensor 2 Image type human sensor 4 Lighting fixture 30 Detection processing part 32 Switch circuit D1 circumscribed rectangle EN entrance / exit Ta start time Tb margin time Td1 initial delay time Td2 exit delay time VF visual field

Claims (4)

A hot-wire sensor that detects the presence of a person by monitoring the heat rays that are installed indoors and radiated from the human body, and that is activated after a person is detected by the heat-ray sensor and takes an indoor image In addition, an image-type human sensor that detects the time when a person in the image leaves the room, a switch circuit that turns on and off the load, a human detection time by a hot-wire human sensor, and a detection time of a room exit by an image human sensor Is the starting point for starting the time limit of the delay time, and during the time of the delay time, the switch circuit must be kept on and both the hot wire type human sensor and the image type human sensor must detect the person within the delay time. if a detection processing unit for turning off the switch circuit in the expiration time of the delay time, hot-wire motion sensor is imaged field and the image type motion sensor for monitoring the heat ray field Are set to be equal, and the detection processing unit starts the delay time that is timed from the time of human detection by the hot-wire human sensor as a constant initial delay time and subtracts a certain margin time from the initial delay time. The time is set, and when the activation time elapses from the time of detection of the person by the heat ray type human sensor, the image type human sensor is activated, and further, a delay time that is timed from the time of detection of leaving the room by the image type human sensor, The exit delay time is shorter than the initial delay time, and during the exit delay time, at least one of the sensitivity of the hot-wire human sensor and the image human sensor is higher than the other periods. Type human touch switch. The image-type human sensor is installed so as to have a downward visual field including at least an entrance and exit, sets a circumscribed rectangle in an area where a person can be considered to exist in the image, and coordinates of a representative point of the circumscribed rectangle in the image are 2. The composite human according to claim 1 , wherein a time when the circumscribed rectangle disappears from the visual field and moves from the central part to the peripheral part of the visual field of the image type human sensor is determined as the time when the person leaves the room. Feeling switch. The image-type human sensor is installed with a visual field obliquely downward toward the entrance and exit, sets a circumscribed rectangle in an area where a person can be considered to exist in the image, and has a vertical dimension and a horizontal dimension of the circumscribed rectangle in the image. claim, at least one of reduced with time, and a time when the area of the point and enclosing rectangle circumscribing rectangle from view disappears becomes below a specified threshold, and determines that exit point of the human 3. A composite human-type switch according to claim 1 or claim 2 . The image-type human sensor sets a circumscribed rectangle in a region where a person can be considered to exist in the image, and the detection processing unit regards the maximum number of circumscribed rectangles generated in the image during the period from entering the room as the number of people entering the room. composite motion switch according to any one of claims 1 to 3, characterized in that the exit number is to detect the time of leaving by the image type motion sensor the matching point to the entry number.

Images