JP2010256045A - Wide range/high accuracy human body detection sensor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a low-cost wide range/high detection accuracy human body detection sensor preventing malfunction without being affected by external disturbance, properly controlling the operation of illumination or air-conditioning or the like immediately after the exit of a person, properly reducing the electric power load for saving energy, and having durability. <P>SOLUTION: The wide range/high definition human body detection sensor is composed of an infrared detection part, a signal processing part processing an analog signal input from the infrared detection part, an arithmetic processing part performing the data processing of the signal input from the signal processing part to determine whether or not a person is present in a detection area, and an input/output part transmitting and receiving information to/from an external device. In the sensor, the infrared detection part includes: a multi-element type thermopile array including three or more elements capable of performing the detection in a unit detention object area at required accuracy; and an assembly infrared light-condensing body integrally including a plurality of unit infrared light-condensing bodies in which light-condensing parts condensing the light to the respective elements so that the entire range of the object area composed of the plurality of unit detection object areas is substantially detectable without omission. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a wide-area, high-accuracy human body detection sensor that detects a stay / absence / entrance / exit of a person in a large space such as an office, a hospital, or a museum.

Conventionally, a human body detection device using a pyroelectric infrared sensor has been widely adopted as a human sensor system that illuminates only when there is a person in order to save energy (Patent Document 1).
The human body detection apparatus includes a plurality of sensor blocks each including one pyroelectric infrared sensor, a reflecting mirror having a plurality of reflecting surfaces formed on the one sensor, and a signal processing unit that determines the presence or absence of a human body. It is provided and covers a plurality of independent detection areas on both sides along the axis, but there are many non-detection areas, and the detection areas cannot be widened and have high detection accuracy.
Further, since the human body detection device is composed of a pyroelectric infrared sensor, it is impossible to detect a stationary human body by detecting only the presence or absence of human movement.
For this reason, it has been simulated as a human body detection signal by holding the detection signal for a certain period of time with a timer.

Many methods for improving the detection accuracy have been proposed (for example, Patent Document 2 and Patent Document 3). Basically, a pyroelectric infrared sensor is used to detect only a human motion detection state. Because the presence of a person is created in a pseudo manner by a timer for a certain period of time, even if a person stays in a stationary state after a set period of time (the stationary state in FIG. 4), it is determined that the person is absent (see FIG. 4). For example, the illumination is turned off (control signal OFF in FIG. 5).
As described above, the human body detection signal is not synchronized with the stay / absence / entrance / exit state of the person in the area, and as a result, as shown in FIGS. Will cause a malfunction.

In addition, since the human body detection signal is output for a certain period of time after the person leaves (stay determination in FIG. 5), the lighting load or the air conditioning operation cannot be appropriately reduced even in the absence state. .
For these reasons, offices, hospitals, etc. have been used only for some toilets, but have not yet spread throughout the buildings of offices, hospitals, etc.

Therefore, in order to solve the above problems of the pyroelectric infrared sensor,
A temperature distribution detecting means in which a thermopile is arranged in a matrix in correspondence with each region in order to divide the living space into a plurality of regions and simultaneously detect the temperature of each region;
Storage means for storing temperature data corresponding to the number of areas detected by the temperature distribution detection means in association with a predetermined area of the divided areas;
Human body detection means for recognizing the presence or absence of a human body existing in the living space based on the temperature distribution detection result of the living space stored in the storage means;
Has been proposed (Patent Document 4).

  When detecting the presence / absence of a human body from a temperature signal detected by a thermopile infrared sensor, this human body detection device increases the number of pixels (temperature signal detection density for a detection target) as in a thermograph, and its image shape A method for discriminating the human body from the above is used.

  However, when this is applied to an office, the elements are closely arranged in a matrix, so that if a human body is detected by a human body detection device over a wide area, a huge number of thermopiles are required and a huge cost is required. It will take.

  Therefore, in order to reduce the number of thermopile usage, a human body detection device has been proposed in which an infrared detection means including a multi-eye thermopile sensor that detects indoor infrared rays is scanned in the horizontal direction by a drive means (Patent Document 5). ).

However, this human body detection device was developed mainly for home air conditioners and is suitable for equipment even if its life is not as long as left, but it is suitable for long-life buildings such as offices and hospitals. It is inappropriate for long-term use in air conditioning systems and lighting systems.
Therefore, it is not practical to employ a mechanism that is always driven in the office from the viewpoints of both life and maintenance of the human body detection device.

  Furthermore, in the case of a technique for discriminating a human body from a change in a temperature detection signal detected by a thermopile infrared sensor (Patent Document 6), a person does not stay for a long time such as an intruder detection use or a crime prevention use, such as sunlight. It is assumed that the environment is free from disturbances and is not suitable for general purpose use.

As described above, since the thermopile infrared sensor has a small detection area, in order to monitor the entire wide area such as an office or a hospital, a mechanism for moving many sensors or automatically moving the sensor is required. It was necessary and it was difficult to adopt from the viewpoint of durability and cost.
For these reasons, human body detection devices have been used only in some places such as toilets and corridors in offices and hospitals, and have not yet spread to buildings such as offices and hospitals.

Japanese Patent Laying-Open No. 2005-084033 Japanese Patent No. 3579907 JP 2005-172377 A Japanese Patent No. 3805165 JP 2008-309379 A Japanese Patent No. 3451238

  The present invention was created to solve the various problems described above, and is not affected by disturbances such as solar radiation, residual heat of the human body, temperature changes of air conditioning, etc., and malfunctions that cause absence determination when a person is stationary for a certain period of time. As soon as people leave the building, they can properly control lighting, air conditioning, etc. to reduce power load appropriately and contribute to energy conservation. Long-lived buildings such as offices and hospitals An object of the present invention is to provide a human body detection sensor having a wide area and high detection accuracy that is suitable for a long-term use in an air conditioning system and a lighting system, and that is low in cost.

  According to the first aspect of the present invention, an infrared detection unit, a signal processing unit that processes an analog signal input from the infrared detection unit, a data input from the signal processing unit, and a person in the detection area In a wide-area, high-definition human body detection sensor composed of an arithmetic processing unit that determines whether or not it exists, and an input / output unit that transmits and receives information to and from an external device, the infrared detection unit requires a unit detection target area. A multi-element type thermopile array having three or more elements that can be detected with high accuracy, and a unit infrared concentrator formed with a condensing part for condensing each of the plurality of elements, In order to be able to detect the entire target area consisting of a plurality of unit target areas almost without any failure, it is composed of a collective infrared condensing body provided integrally with a plurality of unit target areas.

  The invention according to claim 2 is characterized in that the state of the arithmetic processing unit is changed between an absent state, an approach state, a stay state, a human body motion detection state, and a leaving state.

  According to a third aspect of the present invention, when the arithmetic processing unit determines that in the staying state, a person stays in a completely stationary state that does not move at all for a long time in the detection area, it remains in the staying state and is in another state. It is characterized by not making transition to.

  The invention according to claim 4 is configured such that the arithmetic processing unit is not changed to the absence state unless after the transition from the stay state to the human body motion detection state and further to the leaving state. It is characterized by.

  According to a fifth aspect of the present invention, the arithmetic processing unit transitions from the human body motion detection state to the withdrawn state during a procedure of transitioning from the stay state to the human body motion detection state during the stay state procedure. In any of the procedures, the procedure of transitioning from the leaving state to the absent state constitutes an infinite loop that returns to the starting point of the staying state, and the absence of the absence by circulating through any of the infinite loops. It is characterized by not changing to the state.

  The invention according to claim 6 is provided with a number assumption means for assuming the number of people staying in the arithmetic processing unit, and when transitioning from the human body motion detection state to the withdrawal state, the assumed number of people is decreased, and the human body motion detection is performed. When transitioning from a state to the stay state, the assumed number of people is increased.

  In the invention according to claim 7, when the arithmetic processing unit determines that the human body temperature is lower than the ambient temperature / floor temperature, it continuously extracts the entry information and the exit information, and counts the number of times. It is characterized in that a human body in an unsteady state is detected by grasping the number of persons or calculating / monitoring a difference in detection values between elements.

  The invention according to claim 8 further includes a correction temperature sensor for measuring either the surface temperature of the sensor itself or the air temperature around the sensor, and the signal processing unit or the arithmetic processing unit is configured to input the detection signal. Is corrected based on the correction detection signal so as not to be affected by the room temperature.

  The invention according to claim 9 is characterized in that the arithmetic processing unit, the detection value of each element of the infrared detection unit, the fluctuation amount per unit time of the detection value of each element, the increase or decrease of the detection value, the change amount of the detection value, Parameters such as the number of elements in which the detection value changes, the number of elements in which the detection value changes, the difference in the detection value change start time, the difference in the amount of change between each element, the difference in the detection value between each element, etc. By analyzing, heat sources other than humans are detected, or the influence of temperature changes originating from non-humans is removed.

  According to a tenth aspect of the present invention, the arithmetic processing unit filters the detection value of the infrared detection unit with a high-pass filter having an appropriate cut-off frequency and filter order, thereby removing the influence of temperature changes originating from other than humans. It is characterized by extracting information that is necessary or necessary.

According to the first aspect of the present invention, it is possible to prevent the occurrence of a malfunction that makes the absence determination that the person is not present even if the person is stationary for a long time in the detection area. Can be possible.
In addition, immediately after leaving the human detection area, it is possible to appropriately control the transmission of lighting, air conditioning, etc., and to appropriately reduce the power load, thereby contributing to energy saving.
Furthermore, since this wide-area, high-precision human body detection sensor does not have a drive unit, it has durability suitable for long-term use in air-conditioning systems and lighting systems for long-life buildings such as offices and hospitals.
Furthermore, the detection area is broadened so that a relatively small unit detection target area can be detected by a plurality of elements and the entire target area consisting of a plurality of unit target areas can be detected without fail while maintaining high accuracy. In addition, the sensor can be miniaturized and provided at low cost.

According to the inventions according to claims 2 to 5, since the state transition is performed among the absence state, the approach state, the stay state, the human body motion detection state, and the leaving state, the approach state and the leaving state are It is clearly demarcated and cannot transition to an absent state without going through a leaving state.
In other words, the absence determination is not performed unless it is detected that the person has moved out, so that a malfunction that does not exist is not caused although a person is staying, unlike a conventional human body detection sensor.

  According to the sixth aspect of the present invention, since it is possible to assume the approximate number of people staying in the detection area, this assumption result can be applied to control of the air conditioning load.

  According to the invention of claim 7, the number of times of entering / leaving is counted to grasp the approximate number of people, or the difference in the detected value between each element is calculated and monitored to detect a human body in an unsteady state. As a result, the human body can be detected even when the human body temperature is lower than the ambient temperature of the sensor or the floor surface temperature of the detection area, even in a non-steady state.

  According to the eighth aspect of the present invention, since the temperature sensor for correction is further provided and the input detection signal is corrected based on the detection signal for correction, high accuracy of human body detection that is not affected by the indoor temperature at all. Can be achieved.

According to the invention which concerns on Claim 9, the influence of the temperature change by all the heat sources originating in those other than a person can be removed correctly, and the high precision of a human body detection can be achieved.
In addition, when a heat source is detected, it is possible to specify whether the influencing factor for temperature change is the temperature change of the entire indoor environment due to air conditioning, solar radiation, OA equipment, or the residual heat of the human body. Independently of this, the presence / absence information of a person can be acquired.
Therefore, these data processing results can be utilized for air conditioning and lighting control, and appropriate operation control can be performed to contribute to energy saving.

According to the invention which concerns on Claim 10, the influence of the temperature change by all the heat sources originating in those other than a person can be removed correctly, and the high precision of a human body detection can be achieved.
In addition, the entry / exit information can be extracted to roughly determine the number of people, one person, or a plurality of persons.

FIG. 1 is a drawing-substituting photograph showing a wide-area, high-precision human body detection sensor of the present invention. FIG. 2 is a functional block diagram of the wide-area, high-precision human body detection sensor of the present invention. FIG. 3 is a schematic view of a detection area showing that the entire target area of the sensor of the present invention is divided into six unit detection target areas. FIG. 4 is a detection function diagram of the pyroelectric human body detection sensor. FIG. 5 is a diagram showing the relationship between the sensor detection value of the pyroelectric human body detection sensor and the control signal. FIG. 6 is a detection function diagram of the wide-area, high-precision human body detection sensor of the present invention. FIG. 7 is a diagram showing the relationship between the sensor detection value and the control signal of the wide area / high accuracy human body detection sensor of the present invention. FIG. 8 is a state transition diagram showing a state transition between the absence state, the entry state, the stay state, the human body motion detection state, and the leaving state. FIG. 9 is a flowchart illustrating a procedure from the absence state to the entry state. FIG. 10 is a flowchart illustrating a procedure from the entry state to the stay state or the absence state. FIG. 11 is a flowchart illustrating a procedure from the stay state to the transition to the human body motion detection state. FIG. 12 is a flowchart illustrating a procedure from the transition from the human body motion detection state to the withdrawal state or the stay state. FIG. 13 is a flowchart illustrating a procedure from the leaving state to the transition to the absent state or the staying state.

First, with reference to FIGS. 1 to 3, the system configuration and functions of an embodiment of the wide-area, high-precision human body detection sensor of the present invention will be described.
<< System configuration >>
This wide-area, high-accuracy human body detection sensor includes an infrared detector composed of a multi-surface reflecting mirror consisting of two or more surfaces or a multi-surface lens and a thermopile array consisting of three or more elements, a surface temperature of the dust-proof cover and housing, or a sensor. A correction temperature sensor that measures the air temperature around the installation site, a signal processing unit that processes these analog signals, and a human body detection signal and heat source identification signal (analyzed by data analysis using a human body detection algorithm for the office that takes the processed signal) An arithmetic processing unit (CPU) for calculating information such as an OA device discrimination signal) and an input / output unit for transmitting / receiving information to / from an external device.

Referring to FIGS. 1 and 3, the infrared detection unit of the present embodiment is composed of six-element thermopile array circle numbers 1 to 6 and six-surface reflecting mirrors A to F.
The surfaces A to F of the six-surface reflecting mirror correspond to the six unit detection target areas in the zones A to F shown in FIG. 3, and each of the six elements is included in each surface A to F of the reflecting mirror. On the other hand, it has a mirror surface that collects infrared rays.
That is, the detection target area of the present embodiment is divided into six unit detection target areas in zones A to F. Thereby, the detection target area of the human body detection sensor is widened and the detection accuracy is enhanced.
Therefore, although detection signals are input from the six unit detection target areas A to F to each element, the sum of the detection signals from the six unit detection target areas is used as a detection signal for each element thermopile array.
Note that the output level of the detection signal decreases as the temperature increases and the amount of infrared rays increases.
In the present embodiment, the light collector is configured by a six-surface reflecting mirror, but may be configured by a lens that focuses light on six elements.

As shown in FIG. 3, the elements corresponding to the unit detection target areas A to F are composed of two sets of six elements, circle numbers 1 to 3 and circle numbers 4 to 6, and each set of elements. Is composed of the central main element round numbers 2, 5 and the subelement round numbers 1, 3, 4 and 6 on both sides.
In this way, one set of three elements, with the center as the main element and both sides as the sub elements, the element located at the center is highly sensitive, and as can be understood from FIG. This is because the temperature information signal of other areas other than the monitoring area is hardly received.
Therefore, the present invention requires at least three elements.
And when adding this, since 3 elements are performed as 1 set, the element of the multiple of 3 will be provided.
In this embodiment, this means that two sets of elements are provided.
Note that the number of elements in each set need not be an odd number, and can be an even number. In this case, the average of the detection values of the two elements on the center side may be used as the detection value of the main element.

The signal processing unit filters the signal detected by each element of the thermopile array with a hardware filter, receives the signal from the temperature sensor for correction shown in FIG. 2, corrects the detected value of each element, and outputs the output level. It is intended to adjust.
The processing for receiving the signal from the correction temperature sensor and correcting the detection value of each element can also be executed in the arithmetic processing unit.
Therefore, in the following, it is described that correction processing is performed in the arithmetic processing unit.

And the detection signal from a sensor and the signal from the temperature sensor for correction | amendment are input into the said arithmetic processing part (CPU), and the parameter calculated in an arithmetic processing part is as follows.
・ Fluctuation amount per unit time of detection value of each element, that is, slope of detection value ・ Increase / decrease (increase / decrease) in detection value of each element
-Amount of change in the detected value of each element-An element in which the detected value changes (classification of main element or sub-element)
-Number of elements in which the detected value changes-Difference in detection value change start time between each element-Difference in detected value change between each element-Difference in detected value between each element (for example, element 1) And difference between elements 3)

In addition, what is calculated by the CPU and stored in a storage unit (not shown) and used to determine the current state of each detection area is as follows.
・ Absence base value ・ Detected value before human body movement ・ Correction judgment value ・ Human body detected value (used to detect the approximate number of people or one / multiple)
・ Estimated number of people (approximate number or one / multiple detection)
-Residual heat judgment value (used to exclude the effects of residual heat remaining after the human body leaves)

≪System functions≫
As shown in FIGS. 6 and 7, the wide-area / high-precision human body detection sensor of the present invention can continue to detect not only the approaching state and the leaving state but also the stationary staying state.
As a result, it can be determined that the person is staying even if the person does not move at all for a long time in the detection area and remains completely stationary.
And this wide area and high precision human body detection sensor is affected by temperature change in a short time (several minutes to several tens of minutes) by air conditioning control, etc., in the external environment and the whole area for a long time (1 to several hours). Eliminate the effects of temperature changes and solar radiation in the perimeter zone.
For this reason, it is possible to distinguish between the human body and other heat sources such as PC (personal computer), copier, and fax machines, and to distinguish between the human body and the remaining heat of the human body remaining on a chair or desk, etc. / Determination status and human body motion detection status, can detect a stationary human body over a long period of time and detect a completely stationary human body, know whether one or more people stay in the detection area, The human body can be detected in a non-steady state when the temperature is lower than the floor surface temperature.

  As shown in FIG. 8, the wide-area human body detection sensor according to the present invention is characterized in that state transition is performed between an absent state / an approach state / a stay state / a human body motion detection state / a leaving state.

Here, the meaning of each state and the mode of transition from each state will be described.
The absence state means a state in which no human body exists in the detection area, and shifts from the leaving state to the entering state.
The entry state means a state in which a human body has entered the detection area from the absence state, and shifts from the absence state to the stay state or absence state.
The staying state means a state in which a human body stays in the detection area and is stationary or does not move greatly. The staying state shifts from the entry state / withdrawal state / human body motion detection state to the human body motion detection state.
The human body motion detection state means a state in which the human body is in the staying state and is moving, and shifts from the staying state to a leaving state or a staying state.
The leaving state means a state in which the human body that has been in the staying state has left the detection area, and shifts from the human body motion detection state to the staying state or the absence state.

As can be understood from the state transition mode described above, there is no direct state transition from the stay state to the absent state.
In order to make a state transition from the staying state to the absence state, as shown in FIG. 8, the state transition to the absence state is made only after the state transition of the human body motion detection state and the leaving state.
For this reason, even if the person continues in a completely stationary state where the person does not move at all for a long time in the detection area, it can be determined that he / she is staying, in other words, not being absent.
In this respect, if the pyroelectric infrared sensor is stationary for a certain period of time, it is remarkably different from one in which the stay state cannot be determined, and no erroneous determination is made.

Hereinafter, the human body state determination method as a basic algorithm will be described for each state transition mode.
In this state transition main routine, the absent state is the basic state (initial state), and when the state transitions to the absent state, the routine up to that point is temporarily terminated, returning to the beginning of FIG. Cycle through an infinite loop until the value change criterion is met.
Thereafter, when the human body enters again, the value in the absence state immediately before the transition to the entry state is set as a reference value used for each determination.
In the following, it is divided for convenience of explanation, but it should be understood that the actual routine is continuous.

<Transition from absence to entry state (transition)>
Referring to FIG. 9, in the absence state, the CPU satisfies each determination criterion (change is greater than or equal to the threshold value), and each parameter continuously satisfies the determination criterion even after waiting for a certain period of time. If it is determined that the detected value of each parameter in the absence state is stored in the storage unit, the state shifts from the absence state to the entry state.
The parameters used here are the amount of change in the detected value, the element where the change is observed (main element and sub-element), the number of elements where the change is observed, the difference in the change start time between each element, and the difference between each element There are five differences in the amount of change.

<Transition from the entry state to the stay state / absence state (transition)>
As shown in FIG. 10, after the thermopile detection signal is stabilized (the change is within the threshold), the CPU compares the detected value in the absence state read from the storage unit of each parameter with the current detected value. When it is determined that the difference is within the threshold value (judgment reference value), the value of each parameter stored in the storage unit is cleared to the absence state, and is equal to or greater than the threshold value. When it is determined, it shifts to a staying state.

  In this way, when a person once enters the detection area and immediately leaves, the pyroelectric infrared sensor imitates the time stay state set by the timer and continues to illuminate, for example. According to the wide-area, high-precision human body detection sensor, it is possible to immediately determine the absence state, and reduce the load such as lighting and air conditioning, which can contribute to energy saving.

<Transition from stay state to human body motion detection state (transition)>
Referring to FIG. 11, in the staying state, when the CPU determines that the change amount of the detected value of each parameter is equal to or greater than the threshold value, the CPU holds the value of each parameter before the change in the storage unit. Transition to the motion detection state.
On the other hand, when it is determined that the amount of change in the detected value of each parameter is less than or equal to the threshold value in the stay state, the CPU determines whether there is an environmental change effect and whether there is an influence from another heat source. If it is determined that there is an environmental change or the influence of other heat sources, a heat source other than a person is detected and the correction judgment value of each parameter is corrected, and the correction judgment value held in the storage unit is Rewrite and return to the stay state.
That is, when there is no change in the detected value of each parameter in the staying state, it circulates through an infinite loop and does not shift to the human body motion detection state, and hence to the absence state described later.
Therefore, although the person is staying, there is no malfunction that the person is absent.

<Transition from human body motion detection state to stay state / exit state (transition)>
Referring to FIG. 12, after determining that the thermopile detection signal is stable (the change is within the threshold), the CPU reads the value of each parameter read from the storage unit and the current detection value before detecting the human body motion. In comparison, when it is determined that the detected value has increased by a threshold value (judgment reference value) or more, the assumed number of persons is decreased and the process shifts to the leaving state.
When it is determined that the detected value has decreased by more than the threshold value, the assumed number of persons is increased. The assumed number of people will be described in detail later.
In this case, the determination reference value (threshold value) of each parameter is corrected and the state transitions to the stay state, as in the case where it is determined that none of them applies.
Here, it can be understood from FIG. 11 and FIG. 12 that a large infinite loop is formed that transitions from the staying state to the human body motion detection state and returns to the staying state again.

  Even if the CPU shifts from the staying state to the human body motion detection state, it does not shift to the retreating state as long as it does not determine that the detected value has increased beyond the threshold value, so the staying person stays without moving It can return to the state and maintain that state. In this respect, it is greatly different from the conventional pyroelectric human body detection sensor.

<Transition from the leaving state to the staying state / absent state (transition)>
Referring to FIG. 13, this embodiment is a method for determining whether to move from the leaving state to the staying state or the absence state, and includes the following three procedures.
The first procedure is to determine whether or not the difference between the determination reference value in the absence of each parameter read from the storage unit or the corrected determination value obtained by correcting the value and the current detection value is within a threshold value. , Whether each parameter has returned to the value in the absence state,
The second procedure is to determine whether or not the detected value of each parameter satisfies criteria such as the assumed number of people and the difference in the detected value between the elements,
The third procedure is whether there is residual heat from a chair or desk,
Are determined in order, and when it is determined that the exit determination condition of these procedures is satisfied, the storage of each parameter is canceled and the absence state is entered.

  In addition, the CPU determines all of the above-described exit determination conditions. If none of the conditions is satisfied, the CPU determines whether the change in each parameter is equal to or greater than a threshold value. When the determination is made, the assumed number of people is decreased, and when it is determined that the number is equal to or less, the determination reference value of each parameter is corrected without changing the assumed number of people and the state is shifted to the stay state.

≪High-precision human body detection technique≫
As described above, the state determination method of the human body has been described for each state transition mode. However, if the wide-area, high-precision human body detection sensor of the present invention is incorporated into an actual building air conditioning control or lighting control system, the human body It is conceivable that the detection accuracy is insufficient.
Therefore, the present invention provides a method for improving the accuracy of human body detection described below in order to further improve the accuracy of human body detection.
There are the following three techniques for improving the accuracy of human body detection.

The first high accuracy method measures the surface temperature of the dust-proof cover and housing of the human body detection sensor or the air temperature around the sensor installation location by the above-described correction temperature sensor, and the information is subjected to the above signal processing. This is to prevent the influence of temperature change due to air conditioning control or the like by inputting to the CPU via the unit and causing the CPU to correct the thermopile detection value.
This high accuracy method can be applied to exclude the influence of temperature change in a short time (several minutes to several tens of minutes) due to air conditioning control or the like.
A specific application example will be described later.

  The second high-accuracy technique allows the CPU to remove or require the influence of temperature change due to air conditioning control or the like by filtering the sensor detection value with a high-pass filter having an appropriate cutoff frequency and filter order. To extract information.

This high accuracy technique can be applied to various events.
That is, if the cutoff frequency and filter order of the high-pass filter are set according to the target to be removed or extracted and filtered, various influences can be removed or target information can be extracted. .
For example, the influence factors that can be removed include the effect of temperature change in a short time (several minutes to several tens of minutes) due to air conditioning control, etc., and the temperature change in a long time (1 to several hours) due to air conditioning control, etc. There are influences, the influence of other heat sources such as PC / copy / fax office automation equipment and humidifiers, and the residual heat of the human body remaining on chairs and desks.
In addition, information extracted in this embodiment includes entry / exit information.
A specific application example will be described later.

The third high accuracy method is a method for analyzing the parameters of each element of the thermopile array, and it is a large analysis method when a change occurs in all elements and an analysis method when a change occurs in some elements. Separated.
A specific application example will be described later.

≪Example of application of high precision method≫
In the present embodiment, a high accuracy technique is applied for the next information processing.
(1) Excluding the effects of temperature changes in a short time (several minutes to several tens of minutes) due to air conditioning control, etc.
(2) Excluding the effects of temperature changes over a long period (1 to several hours) in the external environment or the entire area,
(3) Exclusion of solar radiation effects in the perimeter zone,
(4) Knowing the number of people staying in the surveillance area (one / multiple),
(5) Discrimination from other heat sources (OA equipment: PC, copy, FAX, humidifier, etc.)
(6) Recognizing the remaining heat of the human body remaining on a chair or desk, and
(7) Non-stationary human body detection, such as during extreme heat and severe winter, when the human body temperature is different from normal (when human body temperature <ambient temperature / floor temperature)
Hereinafter, an algorithm for each technical item will be described.

<Exclusion of effects of temperature changes in a short time>
In order to exclude the influence of the temperature change in a short time, the first to third high accuracy techniques described above were applied.
First, the correction temperature sensor measures the surface temperature of the dust-proof cover of the sensor and the housing, or the air temperature around the sensor installation location, and inputs the information to the CPU via the signal processing unit. Then, by causing the CPU to correct the thermopile detection value, there is no influence of temperature change in a short time (several minutes to several tens of minutes) due to air conditioning control or the like.

  Second, by letting the CPU filter the sensor detection value with a high-pass filter with an appropriate cutoff frequency and filter order corresponding to temperature changes in a short time, a short time (several minutes to several tens of minutes) by air conditioning control, etc. The effect of temperature change in units) is removed.

The third method is an analysis method when all elements are changed in the third high accuracy method.
When the CPU determines that the amount of change per unit time of all the elements of the six-element thermopile array is the same, it can be regarded as an influence due to an environmental change.
Therefore, by causing the CPU to calculate and monitor the difference between the elements, the influence of the temperature change in a short time (several minutes to several tens of minutes) due to air conditioning control or the like is prevented.

<Exclusion of long-term temperature change>
The sensor change value is filtered by the CPU with a high-pass filter with an appropriate cutoff frequency and filter order corresponding to the temperature change over a long period of time, so that the temperature change over a long period of time (1 to several hours) due to air conditioning control, etc. Remove the effects of.

<Exclusion of the effects of solar radiation>
When changes in the detected values due to temperature rise due to solar radiation on the floor surface and sensor body (dust-proof cover / housing) in the perimeter zone appear in all elements, the effect of solar radiation is the same as described in paragraph 0060. This prevents the malfunction that is detected when a person stays due to the temperature rise caused by solar radiation.

On the other hand, when the difference between the elements is calculated, an event (change in the detected value) that occurs in the same manner in both elements is canceled, and an event (change in the detected value) that occurs only in one of the elements can be captured.
Therefore, when the influence of solar radiation is observed on a plurality of elements (5 to 2 elements) that are not all of the six elements of the present embodiment, the difference between the elements of the sub elements is calculated and there is no difference. Since the influence can be eliminated by excluding the change, it is prevented that a person stays due to a temperature rise due to solar radiation.
This corresponds to an analysis method when a change occurs in some elements in the third high accuracy method.

  Also, the probability of sensor placement is assumed to be very low, but if only one element is affected by solar radiation, the slope of the sensor output from the human body should be greater than the slope of the sensor output due to the influence of solar radiation. From the difference between the inclination of the sensor output value when the surface temperature of the floor or desk changes due to the influence of solar radiation and the inclination of the sensor output value when the human body enters, it can be determined whether the human body or solar radiation.

  In any case, only the human body is discriminated by causing the CPU to calculate and monitor the difference or inclination of the detection value of each element, thereby eliminating the influence of solar radiation.

<Understanding the number of people staying in the monitoring area (single / multiple)>
There are two methods for grasping the number of people staying in the monitoring area.
One is to cause the CPU to filter the sensor detection value with a high-pass filter having an appropriate cutoff frequency and filter order corresponding to entry / exit.
Thereby, the entry information and the exit information are extracted, and the number of times is counted to detect whether the number of people in the area is one, or one or more.
The detected number of persons is stored as a human body detection value in the storage unit.

Second, by letting the CPU classify the detected value of each parameter based on a threshold value as to whether it is one or a plurality of persons, it is assumed that a rough number of persons or one / a plurality of persons in the area are calculated.
The CPU stores the estimated number of persons in the storage unit as the assumed number of persons.

The two methods of grasping the number of people staying in the monitoring area described above constitute the number of people assumption means of the present invention.
This subroutine for grasping the number of visitors is incorporated in an appropriate part of the state transition routine for transitioning between the states described above.
Then, when the CPU detects the human body detection value and / or the assumed number of people, the CPU outputs it to the outside from the input / output unit.

<Distinction from other heat sources>
This is a method for distinguishing other heat sources such as PC, copy, FAX and other heat sources such as humidifiers from short- and long-term temperature changes and solar radiation by the air conditioning control described above.
Short- and long-term temperature changes due to air-conditioning control, etc., and temperature changes due to solar radiation change the detection values of all elements in the thermopile, while office equipment such as PCs, copy machines, and fax machines are installed throughout the target area. Since this is rarely performed, the detection values of some elements are changed.
Therefore, when a change appears in the detection values of some elements, it can be determined that the influence is due to other heat sources such as OA equipment.

One of the methods is that the CPU calculates a difference in detection value change start time between elements (whether or not they are simultaneous) or a difference in detection value change amount between elements (whether or not the changes are the same). If it is within the threshold value, it is judged as the influence of the environmental change.
In addition, since the inclination of the sensor output based on the influence of the human body is larger than the inclination of the sensor output based on the influence of other heat sources, the difference between the inclinations of the sensor output values of the two causes the human body or another heat source such as OA equipment. Is determined.
When it is determined that there is an environmental change effect or another heat source effect, the determination value of each parameter is corrected.
These determination results may be output from the input / output unit to the air conditioning control unit and reflected in the air conditioning control.

The second is to let the CPU filter the sensor detection value with a high-pass filter having an appropriate cutoff frequency and filter order corresponding to another heat source.
Thereby, other heat source information is removed, or the delay time until it discriminate | determines from a human body is shortened.

<Determination of the residual heat of the human body>
This is a technique for discriminating the residual heat of the human body remaining on a chair or desk, but there are two techniques.
One is to eliminate the influence of the residual heat of the human body remaining on the chair or desk by causing the CPU to filter the sensor detection value with a high-pass filter having an appropriate cutoff frequency and filter order corresponding to the residual heat of the human body.
As another method, the CPU calculates and monitors the difference between the elements to determine that the heat is not the human body, or shortens the delay time until it is determined that the heat is not the human body, Determine the absence of a person.

<Human body detection during non-stationary>
This is a technique for detecting a human body in a non-stationary state where the human body temperature is lower than the ambient temperature and the floor temperature.
In the normal routine of this embodiment, it is created on the assumption that the human body temperature is higher than the ambient temperature and the floor temperature.
However, it is assumed that the human body temperature is different from the normal time when the human body temperature is higher than the ambient temperature and the floor surface temperature, such as in the extremely hot season and the severe winter season.
In such a case, the sensor detection value is filtered by an appropriate constant / order (high-pass filter) to extract the entry / exit information, and the number of times is counted to determine the approximate number of people. Detects the human body at all times.
Alternatively, as another method, by causing the CPU to calculate and monitor the difference between the elements, the human body in the non-stationary state is detected or the delay time until the non-stationary human body is detected is reduced, and / Confirm the absence state.

Claims (10)

An infrared detection unit, a signal processing unit for processing an analog signal input from the infrared detection unit, and an operation for determining whether a person exists in the detection region by performing data processing on the signal input from the signal processing unit In a wide-area, high-definition human body detection sensor composed of a processing unit and an input / output unit that transmits / receives information to / from an external device,
The infrared detection unit includes a multi-element type thermopile array including three or more elements capable of detecting a unit detection target area with a required accuracy, and a condensing unit for condensing each of the plurality of elements. The unit infrared concentrator formed with a plurality of unit target areas is made up of a plurality of unit target areas, so that the entire target area can be detected without exception. A wide-area, high-precision human body detection sensor.   The wide-area / high-precision human body detection sensor according to claim 1, wherein the arithmetic processing unit performs state transition among an absence state, an entry state, a stay state, a human body motion detection state, and a leaving state. .   In the stay state, the arithmetic processing unit, when it is determined that a person is still in a completely stationary state that does not move at all for a long time in the detection area, stays in the stay state and does not transition to another state. The wide-area, high-precision human body detection sensor according to claim 2, wherein   4. The arithmetic processing unit according to claim 3, wherein after the transition from the staying state to the human body motion detection state, the arithmetic processing unit does not transition to the absent state unless after the transition to the leaving state. The wide-area, high-precision human body detection sensor described.   The arithmetic processing unit is in the stay state procedure, in the procedure to transition from the stay state to the human body motion detection state, in the procedure to transition from the human body motion detection state to the leaving state, from the leaving state In any of the procedures for transitioning to the absent state, an infinite loop returning to the starting point of the staying state is configured, and the transition to any absent loop does not cause transition to the absent state. The wide-area, high-precision human body detection sensor according to claim 3.   The arithmetic processing unit includes a number assumption unit that assumes the number of people who stay, and when transitioning from the human body motion detection state to the withdrawal state, the number of assumed people is decreased and the human body motion detection state transitions to the stay state. 6. The wide-area, high-precision human body detection sensor according to claim 2, wherein the assumed number of persons is increased.   When the arithmetic processing unit determines that the human body temperature is lower than the ambient temperature / floor temperature, it continuously extracts the entry information and the exit information, and counts the number of times to grasp the approximate number of people, or 7. The wide-area, high-precision human body detection sensor according to claim 1, wherein a human body in an unsteady state is detected by calculating and monitoring a difference in detection value between each element. .   A correction temperature sensor that measures either the surface temperature of the sensor itself or the air temperature around the sensor is provided, and the signal processing unit or the arithmetic processing unit determines the input detection signal based on the correction detection signal. The wide-area / high-precision human body detection sensor according to claim 1, wherein the wide-area / high-accuracy human body detection sensor is corrected so as not to be affected by an indoor temperature.   The arithmetic processing unit is configured to change the detection value of each element of the infrared detection unit, the amount of fluctuation of the detection value of each element per unit time, the increase or decrease of the detection value, the change amount of the detection value, or a change in the detection value. By analyzing parameters such as the number of elements in which the detected value changes, the difference in the detection value change start time, the difference in the amount of change between each element, the difference in the detected value between each element, etc. The wide-area, high-precision human body detection sensor according to any one of claims 1 to 7, wherein a heat source is detected or an influence of a temperature change derived from a person other than a person is removed.   The arithmetic processing unit is configured to filter the detection value of the infrared detection unit with a high-pass filter having an appropriate cutoff frequency and filter order, thereby removing or requiring the influence of a temperature change derived from other than a person. The wide-area, high-precision human body detection sensor according to any one of claims 1 to 7, characterized in that