JPH0120476B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a device that detects the number of people from infrared rays emitted by human bodies, and provides a device that can determine the degree of crowding by measuring the number of people.
The present invention can be used, for example, when it is required to grasp the number of people present in a predetermined location such as an underground mall, a station premises, a road intersection, an elevator hall or an escalator entrance, and to know the degree of congestion. It is very effective. For example, recently, elevators have adopted group management control, which selects the optimal car based on the call situation and the positional relationship of each car and assigns it to each hall call, in order to increase operating efficiency and improve service to passengers. ing. in this case,
If we could accurately know not only the call status and car location, but also the number of passengers in each car and the number of waiting customers in each hall, we would avoid assigning nearly full cars to floors with many waiting customers, and elevator This will enable more detailed service to customers and more efficient operation.

Conventionally, a method has been considered to detect the number of people by emitting ultrasonic waves and using the reflected waves (Japanese Patent Publication No. 56-51588), but this method only detects items other than people, such as luggage. In addition, the received wave signal changes greatly depending on the posture and movement of the person (even when a person is standing naturally, each part of the body moves slightly), and the ultrasonic waves are also disturbed by the direction of the wind. . In order to avoid this, a method of averaging the signals has been considered, but in this case, for example, a time longer than the motion cycle is required. Furthermore, the received wave voltage varies greatly depending on where a person stands, causing false alarms. On the other hand, regarding the use of infrared rays, there is a proposal to detect passengers inside the car (Japanese Patent Laid-Open Publication No.
56-161271), this method is aimed at detecting the presence or absence of passengers and cannot detect the number of passengers. A method of using Matsuto switches has also been considered, but in this method, Matsuto switches are placed in multiple subdivisions according to the floor area required for each person, and the number of people is detected based on how many Matsuto switches are operating. However, this requires the construction of a place to place it, and its structure makes it impossible or inappropriate to place it in some places. In addition, a device for detecting the degree of congestion in elevator halls using a television camera has been considered (Special Publication No. 57-30782), but this device stores in advance a pattern when the object for which the degree of congestion is to be detected is not in the video. Then, when an object enters the image, it is detected that the pattern is different from the previously stored pattern, and the degree of congestion is detected by calculating the proportion of the different area in the image. However, this method not only detects objects other than humans (such as trunks), but also requires resetting the reference pattern each time there is a change in the installation location (environment) (such as placing a vase of flowers). Furthermore, the processing system becomes complicated, and it can be said that it lacks versatility. In addition, it is possible to detect the number of people using infrared rays emitted by the human body by using an infrared camera or an infrared vidicon, and it is possible to know the number of people from the output status and changes of each pixel, but the equipment is very complicated and expensive. Become something.

An object of the present invention is to provide a crowding degree detection device that accurately detects the number of people with a simple and inexpensive configuration, and is highly versatile without being limited by the location.

The first feature of the present invention is that it includes a condensing and scanning system that condenses and scans radiant energy from a measurement target in order to capture infrared rays emitted by human bodies and detect the number of people. The second feature of the present invention is to utilize the relationship that the width of the output signal waveform of an infrared detection element (hereinafter simply referred to as a detection element refers to an infrared detection element) is proportional to the number of people being measured. . Furthermore, a third feature is that the field of view can be set arbitrarily by using a signal synchronized with the scanning period. The present invention will be explained below based on the drawings. FIG. 1 is a diagram for explaining the above-mentioned scanning method, where a shows the object plane scanning method and b shows the image plane scanning method. In FIG. 1, 1 is a scanning mirror, 2 is a condensing lens, 3 is a detection element, 4 is an imaging plane, and 5 is infrared rays emitted from a measurement target. In the object plane scanning of a, the detection element is placed on the optical axis or very close to it,
Rotation of the mirror. This method scans by changing the direction of the optical axis by swinging. In this scanning method, the detection element is placed on the optical axis or very close to the optical axis, so the angle of view of the optical system only needs to be small, and good optical performance only needs to be achieved very close to the optical axis. . However, the scanning mirror needs to have a large aperture, and rotating or swinging it involves various difficulties. On the other hand, the image plane scanning method b is a method in which scanning is performed by moving a detection element arranged on the image plane without changing the direction of the optical axis. In this case, the detection element is actually fixed and the scanning mirror placed after the condensing system is rotated. Due to the oscillation, the focused energy is imaged onto the sensing element. This method requires that the entire target scanning width is focused on the image plane, so it requires an optical system with a wide angle of view, and it is also required to have good optical performance even in areas far from the optical axis. However,
Since the scanning mirror can be placed close to the image plane, it can be small and the scanning mechanism can be simplified. In principle, the scanning method of this congestion degree detector may be any method, but an example will be described below in which the image plane scanning method is adopted in order to simplify and downsize the mechanism.

Figure 2 shows that a congestion level detection device 10 is installed with an angle of depression on the ceiling of an elevator hall, and a waiting customer 1
3 shows an example of detecting infrared rays 14 emitted from 3. 12 is an elevator car.
FIG. 3 shows the optical system of the congestion level detection device. 11 is an optical system housing case, 15 is a rotary plane mirror (scanning mirror), and its structure is made of synthetic resin (glass etc. may also be used) with aluminum or chromium, which has a high infrared reflectance, deposited on the surface, and a small motor. 1
6 rotates together with the rotating shaft 16a. Reference numeral 17 denotes a synchronous pulse detection section which generates a signal synchronized with rotation using a photointerrupter and sends it to a synchronous circuit to be described later. Reference numerals 18a to 18e indicate infrared detection elements, in which a plurality of inexpensive pyroelectric detection elements having high quality and stable characteristics are arranged. In order to eliminate the influence of sunlight and illumination light, these elements
Use a window material (not shown) with a cut-on filter that blocks light with a wavelength of 6 μm or less. No. 19 is an infrared lens, which is required to have high transmittance for infrared rays emitted from the human body in the wavelength range of 4.5 to 20 μm. Germanium lenses and silicon lenses are suitable for this purpose. The purpose of using this lens is to ensure the operation of the infrared detector by concentrating infrared rays with the lens, increasing the energy density of the element's receiving surface and increasing the output signal. If there is (several m
(within), depending on the application, it may not be necessary to use a lens (the radiant energy density decreases in inverse proportion to the square of the distance). The important points in an optical system like the one shown in Figure 3 are as follows.

(1) The two-dimensional field of view is covered by rotational scanning of the scanning mirror 15 in the x direction and by a plurality of infrared detection elements 18a to 18e arranged in the y direction. Note that the number of detection elements is not limited to five, and may be selected depending on the resolution of the field of view in the y direction and the degree of accuracy required.

(2) Image plane scanning is performed by the scanning mirror 15, and in order to maximize the field of view in the x direction, the lens 19 and the detection elements 18a to 18e are installed closest to the rotating outer edge of the scanning mirror 15. However, compared to the field of view in the y direction (determined by the position of the sensing element in the y direction, the theoretical limit angle is possible up to nearly 180°), the field of view in the x direction is relatively narrow, determined by the F value (focal length/aperture) of the lens. Becomes a corner.

(3) Because of the image plane scanning method using the scanning mirror 15, as the scanning angle (left and right) increases from the central optical axis of the lens 19, the optical gain ( lens 1
In addition to reducing the effective area of the sensing element 18
(the angle of incidence on a to e increases) occurs,
It becomes zero at the limit of visual field. From there on, the contribution of radiation inside the sensor to the sensing elements 18a to 18e increases. Therefore, the internal temperature of the optical system housing case 11 has an important meaning.

(4) Depending on the setting of the optical gain including the infrared transmittance of the lens 19, the equivalent temperature of the amount of infrared radiation incident on the detection elements 18a to 18e generally differs from the actual temperature (the equivalent temperature is (taken as standard based on internal radiation amount based on sensor inner wall temperature).

(5) Since pyroelectric elements are used as the detection elements 18a to 18e, a feature of their operation and function is that they do not respond to constant incident infrared rays. Therefore, although information on the ambient temperature of the sensing elements 18a to 18e is lost, information on the difference from other radiation, that is, information on the alternating current component, survives through scanning. Therefore, the electric signal waveforms from the sensing elements 18a to 18e reflect the amount of radiation fluctuation based on the inner wall temperature of the sensor, and all voltage levels have no meaning in their absolute values, but are based on the potential difference between them. It has meaning only.

Based on the scanning characteristics of the optical system described above, the scanning mirror 15
The waveform of the amount of infrared radiation incident on the detection elements 18a to 18e, which is obtained as the sensor rotates, will be explained. For convenience, only the detectable field of view of the detection element 18c will be considered here (this is only because the field of view to be detected by the other four detection elements is different, and the functions described below are exactly the same). Figure 4a shows one person near the edge of the visual field, one person at the center of the visual field,
This is an example where there are three targets in total. The dash-dotted line represents the internal radiation of the sensor, which decreases from the field edge toward the center of the field, rises again, and reaches an internal-only radiation level again at the opposite field edge. On the other hand, for the target, as shown by the solid line, it is maximum at the center of the visual field and small near the edges of the visual field. The dashed line represents the optical gain curve. The vertical axis shows the radiation amount as an absolute amount. The sum of radiation from the target and radiation inside the sensor enters the detection element 18c. In reality, radiation from the indoor floor comes in, but we will ignore it here for simplicity. The total radiation dose distribution is shown in Figure 4b. As you can see from the figure, due to the optical gain curve, the radiation from the target is quite small near the edge of the field of view, but on the contrary, the internal radiation is large, so the two are added together, and the peak value of the radiation distribution within the field of view is almost the same. Become. However, as mentioned above, the absolute level in FIG. 4b is not preserved (for example, the internal radiation level varies up and down depending on the waveform), and only the radiation variation is important.

That is, since the output voltage e of the detection element 18c is proportional to the change in the amount of incident radiation W,
If the amount of change is taken based on the waveform in FIG. 4b, the peak value will be approximately constant, and if an appropriate threshold value is provided, relatively uniform pulsing within the visual field will be possible.
As for the signal waveform, due to the resolution of the optical system and the response characteristics of the detection element, the target waveform is considerably blurred and the shape is roughly triangular. However, since the falling time constant is particularly long, it is slightly transformed into an integral wave shape. Therefore, compared to the peak value of the waveform when the target is one person, the peak value of the waveform for two people standing close to each other is approximately twice as high. Since the general shape is triangular, the signal width at that time is also approximately doubled (if the sensing element is of a high-speed response, small resolution type, the peak value remains the same and only the width is doubled). Experiments have shown that when there are three or more people, the peak value is about twice that of one person, and only the width is multiplied by the number of people (three times as much for three people and four times as much for four people). ). In other words, only the width shows a proportional relationship with the number of people. The congestion level detection device according to the present invention utilizes this relationship, and can detect the number of people regardless of the density of the people or their location (as the peak of the signal is not used).
From the results of experiments, it has been found that a large number of people can be detected with high accuracy by determining the number of people from the width of each waveform using the width of the waveform for two targets (in a continuous state) as a reference.

Next, an example of the circuit configuration of the present invention is shown in FIG.
Components that are the same as those in FIG. 3 are designated by the same reference numerals. In FIG. 5, 18a' to 18e' are the output signals of the detection elements 18a to 18e in FIG. (18a' to 18e' are connected to the respective amplifiers). The output of each preamplifier is 5 from 20a to 20e.
It forms a system and is input to the band filter circuit group 21. Here, the low cutoff frequency of the filter can be determined by considering the scanning period of the scanning mirror, and is preferably about 0.1 Hz (the higher the chopping frequency, the lower the response sensitivity of the detection element). Further, the high cutoff frequency may be set to approximately 40 Hz in order to remove noise components caused by the commercial power frequency. Outputs 21a to 21e from the 5-system band filter 21
is input to the main amplifier circuit group (5 systems) 22.
The configuration of the main amplifier circuit group 22 is as shown in FIG. 6, and there are five systems, but for the sake of simplicity, only the bandpass filter output 21c will be considered for one system (signal from FIG. 3 18c). 31 is an amplifier, 31c is its output,
A comparator 32 outputs a pulse signal 32c with a signal higher than a predetermined threshold level. 33 is an AND circuit which can be detected only during the time of a wind pulse 26a (details will be described later) and outputs an output 22c. FIG. 7 shows the waveforms and timing of each part in FIG. 6. Figure 7a is 31
If there is no person within the detection field of view, the waveform will be as shown by the broken line in the figure. Figure b shows a signal waveform 32c, in which the comparator 32 picks out only signals above the threshold level TL, resulting in pulses of positive logic level. Figure c shows wind pulse 26
This shows the waveform and timing of a, which is created by the synchronous circuit 26 in Figure 5 (details will be described later). OT shows the opening timing and CT shows the closing timing, and these are variable. The time between and CT is the time width of this channel. By providing this wind pulse 26a, the field of view of the detector can be adjusted. Furthermore, malfunctions due to fluctuations in the internal temperature of the optical system housing case can be avoided. The example in Figure 7 is the waveform when there are four people in the visual field, b' is two people (people in a row), and b'' and b are the waveforms when there are one person each.In this way, the main amplifier The circuit group 22 is a circuit that extracts only a signal (signal corresponding to a person within the field of view) having a predetermined value or more within the field of view set by the wind pulse from the output signal of the detection element.

Returning to FIG. 5, 23 is a digital multiplexer circuit which outputs five systems of pulse inputs as one system of pulse trains. This digital multiplexer circuit 23 measures the width of the pulses as shown in FIG. The output 23a is taken out as a column. Figure 8 shows the digital multiplexer circuit 23.
The configuration is shown below. In FIG. 8, an astable multivibrator 31 generates a clock pulse 31a, which is further frequency-divided by a decimal ring counter 32 to generate five groups of clock pulses 32a to 32e shifted by 1/5 period. 33-37 are
NAND gate circuit, 43 to 47 are diodes,
50 is an inverter, and P is a positive power supply. Therefore, when a pulse is input to 22a to 22e, a sequence of the number of clock pulses corresponding to the width of the pulse can be taken out as output 23a.

Returning to FIG. 5, reference numeral 26 is a synchronous circuit, which provides the operation timing of the main amplifier circuit 22 and the counting circuit 24, and outputs 26a, 26b, 2, respectively.
Controlled by 6c. Synchronous circuit 26
The configuration is shown in Figure 9, and the timing chart is shown in Figure 1.
Shown in Figure 0. 17a in Figures 9 and 10
is the synchronization signal (timing pulse) taken out by the photo interrupter 17 shown in FIG.
One pulse is generated each time the scanning mirror 15 rotates once, and its waveform is shown in FIG. 10a. 26b is a reset pulse obtained by the timing pulse 17a via the delay circuit 65, which is used to reset the contents of the counter in the counting circuit 24;
The waveform is shown in FIG. 10b. 61a is a start pulse obtained via the monostable multivibrator 61, the waveform of which is shown in FIG. 10c, but the time width T can be set arbitrarily. 62a is a pulse for determining the opening timing of the wind pulse 26a, and 63a is a pulse for determining the opening timing of the wind pulse 26a.
These pulses are used to determine the closing timing of pulse a, and are obtained from the start pulse 61a via monostable multivibrators 62 and 63, respectively, and their waveforms are shown in FIGS. 10d and 10e. 64 is
It consists of a NOT circuit and an AND circuit, and the pulse 62a
is input to the NOT circuit, and this output and pulse 63a
By combining them using an AND circuit, a wind pulse 26a can be obtained as shown in FIG. 10f.
As described above, by providing the wind pulse 26a, the detection field of view can be limited, and the temperature inside the optical system housing case (back side part) is not detected, so malfunctions caused by this can be avoided. The timing pulse 26c is the same as the timing pulse 17a and is sent to the counting circuit 24, and the contents of the counter are latched until the next pulse. FIG. 10g shows the waveform of the output 22c of the main amplifier circuit group 22 when there are four people within the detection field of the detection element 18c as explained in FIG. FIG. 10h shows the waveform of the clock pulse 31a in FIG.
m is a clock pulse train frequency-divided by a decimal ring counter 32, each having a waveform of 32a to 32e. Therefore, Figure 10 n is 22c of g and 3 of k
This shows the waveform of FIG. 8 23a synthesized from signals 2c, and it is possible to obtain a pulse train corresponding to the pulse width of 22c, that is, the number of people within the visual field.

Returning to FIG. 5, 24 is a counting circuit composed of a counter. In this circuit, a sequence of pulse numbers inputted by a signal 23a is counted each time the scanning mirror rotates once, and a signal 24 representing the number of people in the field of view is counted.
Output a. In this way, the main amplifier circuit group 22
The digital multiplexer circuit 23 converts the signal corresponding to the person in the visual field into the number of clock pulses corresponding to the width of the waveform, and the counting circuit 24 counts this clock pulse number to calculate the number of people in the visual field or crowding. degree can be determined. 25 is an output device. Output devices include a number of people display and a control signal device that outputs signals according to the degree of crowding. To display the number of people, for example, a 7-segment LED may be driven by a decoder (omitted). Further, a control signal output according to the degree of congestion can be obtained using, for example, a flip-flop circuit (omitted).

In the above embodiment, the field of view of each detection element in the x direction in FIG. 3 differs depending on the characteristics of the lens system (18a and 18e are the minimum, and 18c is the maximum). Therefore, if the wind pulse is made variable for each channel, unnecessary areas can be cut out, so the field of view for each channel can be made clear, and the influence of temperature fluctuations on the inner surface of the case can be avoided (especially on the sensing elements at both ends). Furthermore, by making the wind pulse variable, the detection field of view can be set arbitrarily, increasing the degree of freedom in the installation environment.

Moreover, if the threshold level of the comparator shown in FIG. 632 can be automatically adjusted according to the ambient temperature, sensitivity unevenness due to seasonal temperature fluctuations can be suppressed. To achieve this, a temperature sensor can be built into the optical system housing case (a thermistor or an IC as a temperature sensor controller), and the threshold level can be set using the temperature-related fluctuations in the output voltage ( (circuit omitted).

Furthermore, in the above embodiment, the coating on the back surface of the scanning mirror was applied only to one side of the flat plate.
The coating may be applied to both sides of the flat plate,
In this case, the synchronization signal extracted by the photointerrupter only needs to generate two pulses each time the scanning mirror rotates once. In this case, the rotational speed of the motor may be reduced to 1/2, which is effective in reducing rotational noise. Further, the shape of the scanning mirror is not limited to a flat surface, but may be a polygonal column (polygon mirror) or a curved surface.

FIG. 11 shows another embodiment of the scanning mirror,
In this example, the scanning mirror 70 includes a coating surface 70a parallel to the rotation axis 71 and a coating surface 70b oblique to the rotation axis 71. With such a configuration, the output signal from the detection element obtained by parallel scanning of the coating surface 70a (parallel to the x-axis in FIG. 3) and the oblique scanning of the coating surface 70b (parallel to the x-axis in FIG. 3) By combining the output signals from the detection elements obtained with the oblique movement), the resolution can be further increased, and the presence of a field of view (dead zone) where detection is insufficient due to gaps between detection elements can be eliminated. Can be done.

As described above, the present invention can be constructed at a much lower cost than using an infrared camera or infrared vision, and can detect the number of people with high accuracy. It can be very effective in controlling the

[Brief explanation of drawings]

Figures 1a and b are diagrams for explaining the scanning method of the congestion level detection device according to the present invention, Figure 2 is a sectional view of the elevator hall, and Figure 3 is the configuration of the optical system of the congestion level detection device according to the present invention. Figure 4 a, b
5 is a diagram showing the distribution of the amount of infrared radiation incident on the detection element, FIG. 5 is a diagram showing the circuit configuration of the present invention, FIG. 6 is a diagram showing the circuit configuration of the main amplifier circuit group, and FIG. 7a
~c is a diagram explaining how to obtain a signal corresponding to the number of people in the visual field from the output of the detection element, Figure 8 is a diagram showing the circuit configuration of a digital multiplexer circuit, and Figure 9 is a diagram showing the circuit configuration of a synchronous circuit. , FIG. 10 is a time chart of each signal, and FIG. 11 is a diagram showing another embodiment of the scanning mirror. 1, 15... Scanning mirror, 2, 19... Lens,
3, 18a to 18e...detection element, 5, 14...
Infrared rays, 13...waiting customers, 20...preamplifier group,
21...band filter circuit group, 22...main amplifier circuit group, 23...digital multiplexer circuit, 24...counting circuit, 25...output device, 26
...Synchronous circuit.

Claims (1)

[Claims] 1. A scanning mirror that rotates at a constant cycle, an infrared detection element that outputs a signal according to a change in infrared light incident through the scanning mirror, and a field of view of the infrared detection element that is set in synchronization with the rotation of the scanning mirror. a synchronization circuit that generates a wind pulse to detect the infrared detection element; a circuit that extracts only a signal that is equal to or greater than a predetermined value within a field of view set by the wind pulse from the output signal of the infrared detection element; A congestion degree detection device comprising a circuit that determines the number of people or the degree of crowding within the field of view based on the width.