JP5875541B2 - Entrance / exit management device and program - Google Patents

Entrance / exit management device and program Download PDF

Info

Publication number
JP5875541B2
JP5875541B2 JP2013027128A JP2013027128A JP5875541B2 JP 5875541 B2 JP5875541 B2 JP 5875541B2 JP 2013027128 A JP2013027128 A JP 2013027128A JP 2013027128 A JP2013027128 A JP 2013027128A JP 5875541 B2 JP5875541 B2 JP 5875541B2
Authority
JP
Japan
Prior art keywords
room
people
door
leaving
opening
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
JP2013027128A
Other languages
Japanese (ja)
Other versions
JP2014157422A (en
Inventor
宗佑 南田
宗佑 南田
博志 嘉藤
博志 嘉藤
健次郎 三浦
健次郎 三浦
啓太 齊藤
啓太 齊藤
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Mitsubishi Electric Corp
Mitsubishi Electric Building Techno Service Co Ltd
Original Assignee
Mitsubishi Electric Corp
Mitsubishi Electric Building Techno Service Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Mitsubishi Electric Corp, Mitsubishi Electric Building Techno Service Co Ltd filed Critical Mitsubishi Electric Corp
Priority to JP2013027128A priority Critical patent/JP5875541B2/en
Publication of JP2014157422A publication Critical patent/JP2014157422A/en
Application granted granted Critical
Publication of JP5875541B2 publication Critical patent/JP5875541B2/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Alarm Systems (AREA)

Description

本発明は、室内の在室人数を求める入退室管理装置及びプログラムに関する。   The present invention relates to an entrance / exit management device and a program for determining the number of people in a room.

ビル等の建物を管理するシステムでは、省エネルギー分析に活用する目的で、各部屋の在室人数を把握したい場合がある。例えば、(1)消費電力を在室人数で除算し、人数当たりの消費電力を比較することで、電力消費量が相対的に多い箇所を特定する、(2)在室人数から熱負荷の算出及び換気量の調整を行うことで空調電力を削減する、(3)入退照明連携(在室状況に応じて人のいる場所のみ照明をつける処理)の効果を見積もり、客先への提案に利用する等の目的で、在室人数を把握したい場合がある。   In a system for managing buildings such as buildings, there is a case where it is desired to grasp the number of people in each room for the purpose of utilizing for energy saving analysis. For example, (1) Divide the power consumption by the number of people in the room and compare the power consumption per person to identify locations where the power consumption is relatively high. (2) Calculate the heat load from the number of people in the room And reduce the air conditioning power by adjusting the ventilation volume. (3) Estimate the effect of entrance / exit lighting cooperation (processing to turn on the lighting only in places where people are present according to the occupancy status), and propose to customers You may want to know the number of people in the room for the purpose of use.

従来の入退室管理システムとして、対象となる部屋の入室側(外側)及び退室側(内側)のそれぞれに認証装置を設置することで在室人数を把握するシステムが知られている。しかしながら、多くのビルでは認証装置は部屋の入室側のみに設置されており、部屋から退室する人を検知することができないため、在室人数を正確に把握することは困難である。そこで、追加のセンサやカメラを用いることによって通行人数を把握する方法が考えられるが、センサ等を追加することでコストが増大してしまう。   As a conventional entrance / exit management system, there is known a system for determining the number of people in a room by installing an authentication device on each of the entrance side (outside) and exit side (inside) of a target room. However, in many buildings, the authentication apparatus is installed only on the entrance side of the room, and it is difficult to accurately detect the number of people in the room because it is impossible to detect the person leaving the room. Therefore, a method of grasping the number of passing persons by using an additional sensor or camera can be considered, but adding the sensor or the like increases the cost.

また、特許文献1には、入室カード読取装置の信号が無い状態において、ドアスイッチがオフ状態で、かつ、入退室センサによって退室者の通過を検出することで、退室者か否かを判断する入退室管理装置が開示されている。   Further, in Patent Document 1, it is determined whether or not the person has left the room by detecting the passage of the person leaving the room with the entrance / exit sensor while the door switch is in an off state in the absence of a signal from the room card reader. An entrance / exit management device is disclosed.

特許文献2には、入退室方向に沿った2つのセンサにより入室及び退室を検知する入退室管理装置が開示されている。   Patent Literature 2 discloses an entrance / exit management device that detects entrance and exit by two sensors along the entrance / exit direction.

特許文献3には、扉の開閉時間を計測し、利用者毎の履歴を蓄積し、共連れ等により入退室履歴の欠損を補完する入退室管理装置が開示されている。   Patent Document 3 discloses an entrance / exit management device that measures the opening / closing time of a door, accumulates a history for each user, and compensates for a lack of entry / exit history by sharing.

特許文献4には、複数の部屋を囲む領域の出口に認証装置を設けることで、退室側の認証装置の数を削減しつつ、在室人数を把握するようにした方法が開示されている。   Patent Document 4 discloses a method in which an authentication device is provided at an exit of an area surrounding a plurality of rooms, thereby reducing the number of authentication devices on the exit side and grasping the number of people in the room.

特許文献5には、扉が開いている時間が閾値よりも長い場合、在室人数を1人増加又は減少させるシステムが開示されている。   Patent Document 5 discloses a system that increases or decreases the number of people in a room when the time during which the door is open is longer than a threshold value.

特開平5−35952号公報JP-A-5-35952 特開平6−251219号公報JP-A-6-251219 特開2008−225938号公報JP 2008-225938 A 特開2002−73778号公報JP 2002-73778 A 特開2011−144505号公報JP 2011-144505 A

上述したいずれのシステムにおいても、在室人数を把握するためには、退室側に認証装置を設置したり、センサやカメラを設置したりする必要があるため、コストを増大させずに在室人数を把握することは困難である。例えば特許文献1,2に記載の発明ではセンサを用いる必要があるため、コストが増大する。特許文献3に記載の発明では、入室側及び退室側のそれぞれに認証装置を設ける必要があるため、入室側のみに認証装置が設置されている部屋については在室人数を把握することが困難となる。特許文献4に記載の発明においても、対象領域の出口に認証装置を設置する必要があるため、入室側のみに認証装置が設置されている部屋については在室人数を把握することが困難となる。また、退室側に認証装置が設置されていない部屋について、特許文献3,4に記載の発明を適用しようとする場合、当該部屋の退室側に認証装置を追加で設置する必要があるため、その分、コストが増大してしまう。特許文献5に記載の発明では、扉が開いている時間と閾値とを比較して在室人数を増減しているに過ぎず、正確に在室人数を把握することは困難である。   In any of the systems described above, in order to grasp the number of people in the room, it is necessary to install an authentication device on the exit side, or install a sensor or camera. It is difficult to grasp. For example, in the inventions described in Patent Documents 1 and 2, since it is necessary to use a sensor, the cost increases. In the invention described in Patent Document 3, since it is necessary to provide an authentication device on each of the entrance side and the exit side, it is difficult to grasp the number of people in the room where the authentication device is installed only on the entrance side. Become. Also in the invention described in Patent Document 4, since it is necessary to install an authentication device at the exit of the target area, it is difficult to grasp the number of people in the room where the authentication device is installed only on the entrance side. . In addition, when applying the inventions described in Patent Documents 3 and 4 to a room where no authentication device is installed on the exit side, it is necessary to additionally install an authentication device on the exit side of the room. The cost will increase. In the invention described in Patent Document 5, the number of people in the room is merely increased or decreased by comparing the time when the door is open and a threshold value, and it is difficult to accurately grasp the number of people in the room.

本発明の目的は、部屋から退室する人を検知する装置が設置されていない場合であっても、当該部屋の在室人数を推定することが可能な入退室管理装置及びプログラムを提供することである。   An object of the present invention is to provide an entry / exit management device and a program capable of estimating the number of people in a room even when a device for detecting a person leaving the room is not installed. is there.

請求項1に記載の発明は、部屋に入室する人を検知する検知手段と、前記部屋からの退室時に前記部屋の扉が開いている時間間隔に基づき、前記扉の各開閉時点における退室人数を推定し、前記検知手段によって検知された入室人数と前記推定された退室人数とに基づき、前記扉の各開閉時点における前記部屋の在室人数を求める在室人数推定手段と、を有し、前記在室人数推定手段は、所定期間内の退室人数の合計が、前記所定期間内の入室人数の合計と等しくなるように、退室人数の合計を前記扉の各開閉時点における退室人数に割り振り、退室人数の割り振りが異なる複数の組み合わせのそれぞれについて、前記扉の各開閉時点における入室人数と退室人数とから前記扉の各開閉時点における通行人数を求め、前記扉の各開閉時点における通行人数と、前記扉の各開閉時点において前記扉が開いている時間間隔とに基づいて尤度を求め、尤度が最大となる組み合わせの入室人数と退室人数とから求められる在室人数を推定値として出力する、ことを特徴とする入退室管理装置である。 According to the first aspect of the present invention, the number of people leaving the room at each opening and closing time of the door is determined based on detection means for detecting a person entering the room and a time interval during which the door of the room is open when the person leaves the room. estimated, based on the exit number, which is the presumed entry number detected by said detecting means, have a, and the number of people estimating means for obtaining the number of people of the room at each opening and closing time of the door, the The occupancy estimation means assigns the total number of people leaving to the number of people leaving at each opening and closing of the door so that the total number of people leaving in the predetermined period is equal to the total number of people entering in the predetermined period. For each of a plurality of combinations with different allocations of persons, the number of people passing at each opening and closing time of the door is determined from the number of people entering and leaving the door at each opening and closing time of the door, and each time the door is opened and closed. The likelihood is calculated based on the number of people passing by and the time interval at which the door is open at each opening and closing time of the door, and the number of people in the room determined from the number of people entering and leaving the combination with the maximum likelihood. It is an entrance / exit management device characterized in that it is output as an estimated value .

請求項に記載の発明は、請求項に記載の入退室管理装置であって、前記検知手段は、前記部屋に入室する人を認証して入室人数を検知し、前記在室人数推定手段は、前記複数の組み合わせのそれぞれについて、入室人数と割り振られた退室人数とから前記扉の各開閉時点における在室人数を求め、前記検知手段の認証結果に基づき、同一人物の重複の無い入室人数の合計を前記複数の組み合わせのそれぞれについて求め、前記複数の組み合わせのうち、前記扉の各開閉時点における在室人数が、同一人物の重複の無い入室人数の合計以下となる組み合わせを対象にして、尤度が最大となる組み合わせを決定し、前記決定された組み合わせの在室人数を推定値として出力する、ことを特徴とする。 Invention of Claim 2 is the entrance / exit management apparatus of Claim 1 , Comprising: The said detection means authenticates the person who enters the said room, detects the number of people entering, and the said number-of-rooms estimation means For each of the plurality of combinations, the number of people in the room at each opening and closing time of the door is determined from the number of people entering and the number of people leaving the room, and the number of people entering the room without duplication of the same person based on the authentication result of the detection means For each of the plurality of combinations, among the plurality of combinations, the number of people in the room at each opening and closing time of the door is a combination that is less than or equal to the total number of people entering the room without duplication of the same person, The combination having the maximum likelihood is determined, and the number of people in the room of the determined combination is output as an estimated value.

請求項に記載の発明は、部屋に入室する人を検知する検知手段と、前記部屋からの退室時に前記部屋の扉が開いている時間間隔に基づき、前記扉の各開閉時点における退室人数を推定し、前記検知手段によって検知された入室人数と前記推定された退室人数とに基づき、前記扉の各開閉時点における前記部屋の在室人数を求める在室人数推定手段と、を有し、前記在室人数推定手段は、前記扉の各開閉時点において扉が開いている時間間隔と通行人数とに基づいて求められる尤度が最大となるように、前記扉の各開閉時点における退室人数を決定し、入室人数と決定された退室人数とから求められる在室人数を推定値として出力する、ことを特徴とする入退室管理装置であるThe invention according to claim 3 is based on detection means for detecting a person entering the room, and the time interval at which the door of the room is open when leaving the room. Occupancy number estimation means for determining the number of people in the room at each opening and closing time of the door based on the estimated number of people entering and the estimated number of people leaving the room detected by the detection means, The occupancy estimation means determines the number of people leaving the room at each opening and closing time of the door so that the likelihood obtained based on the time interval at which the door is open and the number of passing persons at the opening and closing time of the door is maximized. In addition, the present invention is an entry / exit management device characterized in that the number of people in the room calculated from the number of people entering the room and the number of people leaving the room is output as an estimated value.

請求項に記載の発明は、請求項から請求項のいずれか一項に記載の入退室管理装置であって、前記在室人数推定手段は、人が前記扉の通行に要する時間の長さに通行人数を乗算して得られる値と、前記扉の開閉に要する時間との合計を、前記扉が開いている時間間隔と定義する通行モデルを用いて、尤度を求める、ことを特徴とする。 The invention described in claim 4 is the entry and exit management apparatus according to any one of claims 1 to 3, wherein the number of people estimating means, people of the time required for passage of the door The likelihood is obtained using a traffic model that defines the sum of the value obtained by multiplying the length by the number of people and the time required to open and close the door as the time interval during which the door is open. Features.

請求項に記載の発明は、請求項に記載の入退室管理装置であって、前記在室人数推定手段は、通行人数と前記扉が開いている時間間隔とに基づいて、人が前記扉の通行に要する時間と前記扉の開閉に要する時間とを新たに求め、前記新たに求められた値と前記通行モデルとを用いて、尤度を求める、ことを特徴とする。 Invention of Claim 5 is the entrance / exit management apparatus of Claim 4 , Comprising: The said occupancy number estimation means is based on the number of passing persons and the time interval when the said door is open, and a person A time required for door passage and a time required for opening and closing the door are newly obtained, and the likelihood is obtained using the newly obtained value and the passage model.

請求項に記載の発明は、請求項から請求項のいずれか一項に記載の入退室管理装置であって、前記在室人数推定手段は、入室人数及び退室人数を増加させ、増加後の入室人数と退室人数とに基づき通行人数を求め、増加後の通行人数と前記扉が開いている時間間隔とに基づいて尤度を求め、増加後の尤度が増加前の尤度よりも高い場合、増加後の入室人数と退室人数とから求められる在室人数を推定値として出力する、ことを特徴とする。 The invention described in claim 6, a entry and exit management apparatus according to any one of claims 1 to 4, wherein the number of people estimating means increases the entry number and exit number, increased The number of people who pass and the number of people who leave the room is obtained, and the likelihood is obtained based on the number of people who have passed and the time interval when the door is open, and the likelihood after the increase is more than the likelihood before the increase. If the number is too high, the number of people in the room calculated from the number of people entering and leaving after the increase is output as an estimated value.

請求項に記載の発明は、コンピュータに、部屋に入室する人を検知するステップと、前記部屋からの退室時に前記部屋の扉が開いている時間間隔に基づき、前記扉の各開閉時点における退室人数を推定し、検知された入室人数と前記推定された退室人数とに基づき、前記扉の各開閉時点における前記部屋の在室人数を求めるステップと、を実行させ、前記在室人数を求めるステップでは、所定期間内の退室人数の合計が、前記所定期間内の入室人数の合計と等しくなるように、退室人数の合計を前記扉の各開閉時点における退室人数に割り振り、退室人数の割り振りが異なる複数の組み合わせのそれぞれについて、前記扉の各開閉時点における入室人数と退室人数とから前記扉の各開閉時点における通行人数を求め、前記扉の各開閉時点における通行人数と、前記扉の各開閉時点において前記扉が開いている時間間隔とに基づいて尤度を求め、尤度が最大となる組み合わせの入室人数と退室人数とから求められる在室人数を推定値として出力する、ことを特徴とするプログラムである。
請求項8に記載の発明は、コンピュータに、部屋に入室する人を検知するステップと、前記部屋からの退室時に前記部屋の扉が開いている時間間隔に基づき、前記扉の各開閉時点における退室人数を推定し、検知された入室人数と前記推定された退室人数とに基づき、前記扉の各開閉時点における前記部屋の在室人数を求めるステップと、を実行させ、前記在室人数を求めるステップでは、前記扉の各開閉時点において扉が開いている時間間隔と通行人数とに基づいて求められる尤度が最大となるように、前記扉の各開閉時点における退室人数を決定し、入室人数と決定された退室人数とから求められる在室人数を推定値として出力する、ことを特徴とするプログラムである。
According to a seventh aspect of the present invention, the computer detects a person entering the room and the exit time at each opening and closing time of the door based on the time interval when the door of the room is open when leaving the room. estimating the number of people, based on the detected entry number and the estimated exit number, determining a number of people of the room at each opening and closing time of the door, to the execution, determining the number of people step Then, the total number of people leaving the room is allocated to the number of people leaving the room at each opening and closing time of the door so that the total number of people leaving the room during the predetermined period is equal to the total number of people entering the room during the predetermined period. For each of a plurality of combinations, the number of people passing at each opening and closing time of the door is determined from the number of people entering and leaving the door at each opening and closing time of the door. The likelihood is calculated based on the number of people passing through and the time interval at which the door is open at each opening and closing time of the door, and the number of people in the room determined from the number of people entering and leaving the combination with the maximum likelihood. It is a program characterized by outputting as an estimated value .
According to an eighth aspect of the present invention, on the basis of the step of detecting a person entering the room in the computer and the time interval at which the door of the room is open when leaving the room, the exit at each opening and closing time of the door Estimating the number of people, obtaining the number of people in the room at each opening and closing time of the door based on the detected number of people entering the room and the estimated number of people leaving, and obtaining the number of people in the room Then, the number of persons leaving the room at each opening and closing time of the door is determined so that the likelihood obtained based on the time interval and the number of passing persons at the time of opening and closing of the door is maximized. The program is characterized in that the number of people in the room determined from the determined number of people leaving the room is output as an estimated value.

本発明によると、部屋の扉が開いている時間間隔に基づいて退室人数を推定することで、部屋から退室する人を検知する装置が設置されていない場合であっても、当該部屋の在室人数を推定することが可能となる。   According to the present invention, even if a device for detecting a person leaving the room is not installed by estimating the number of people leaving the room based on the time interval when the door of the room is open, It is possible to estimate the number of people.

本発明の実施形態に係る入退室管理装置の一例を示すブロック図である。It is a block diagram which shows an example of the entrance / exit management apparatus which concerns on embodiment of this invention. 本実施形態に係る入退室管理装置による動作の一例を示すフローチャートである。It is a flowchart which shows an example of the operation | movement by the entrance / exit management apparatus which concerns on this embodiment. 入退室履歴を示す表である。It is a table | surface which shows an entrance / exit log | history. 動作例を示す表である。It is a table | surface which shows an operation example. 在室人数の推移を示すグラフである。It is a graph which shows transition of the number of people in a room. 変形例1に係る入退室管理装置による動作の一例を示すフローチャートである。It is a flowchart which shows an example of operation | movement by the entrance / exit management apparatus which concerns on the modification 1. 変形例1に係る動作例を示す表である。10 is a table showing an operation example according to Modification 1.

図1に、本発明の実施形態に係る入退室管理装置の一例を示す。本実施形態に係る入退室管理装置10は、入退室管理部12と在室人数推定部14と出力部16とを備え、部屋20への人の入室を管理し、部屋20から退室する人数を推定し、部屋20の在室人数を求める。   FIG. 1 shows an example of an entrance / exit management device according to an embodiment of the present invention. The entrance / exit management device 10 according to this embodiment includes an entrance / exit management unit 12, a occupancy number estimation unit 14, and an output unit 16, manages the entrance of people into the room 20, and determines the number of people leaving the room 20. Estimate and determine the number of people in the room 20.

部屋20は、例えばオフィスビル等の執務室や会議室等である。部屋20には、人の出入口となる扉22が設けられている。扉22には電気錠24が備え付けられている。電気錠24は、解錠信号が入力されると解錠され、施錠信号が入力されると施錠される。また、部屋20から人が退室する場合、電気錠24は解錠される。なお、電気錠24は、解錠されたときに解錠信号を入退室管理部12に出力し、施錠されたときに施錠信号を入退室管理部12に出力してもよい。   The room 20 is, for example, an office room such as an office building or a meeting room. The room 20 is provided with a door 22 serving as a doorway for people. The door 22 is provided with an electric lock 24. The electric lock 24 is unlocked when an unlocking signal is input, and is locked when a locking signal is input. When a person leaves the room 20, the electric lock 24 is unlocked. The electric lock 24 may output an unlock signal to the entrance / exit management unit 12 when unlocked, and may output a lock signal to the entrance / exit management unit 12 when locked.

扉22の付近には、扉22の開閉を検知する扉開閉センサ26が設置されている。扉開閉センサ26は、扉22が開いたことを検知すると、扉22が開いたことを示す扉開信号を入退室管理部12に出力し、扉22が閉じたことを検知すると、扉22が閉じたことを示す扉閉信号を入退室管理部12に出力する。扉開信号には、扉22が開いた時刻(以下、「扉開時刻」と称する)を示す扉開時間情報が含まれ、扉閉信号には、扉22が閉じた時刻(以下、「扉閉時刻」と称する)を示す扉閉時間情報が含まれてもよい。   A door opening / closing sensor 26 that detects opening / closing of the door 22 is installed in the vicinity of the door 22. When the door opening / closing sensor 26 detects that the door 22 is opened, the door opening / closing sensor 26 outputs a door opening signal indicating that the door 22 is opened to the entrance / exit management unit 12. A door closing signal indicating that the door has been closed is output to the entrance / exit management unit 12. The door opening signal includes door opening time information indicating the time when the door 22 is opened (hereinafter referred to as “door opening time”), and the door closing signal is included in the time when the door 22 is closed (hereinafter referred to as “door”). Door closing time information indicating "closed time") may be included.

また、部屋20の外側であって扉22の付近に、認証装置28が設置されている。このように、部屋20の入室側に認証装置28が設置されており、入室時に認証処理が行われ、認証された人が入室を許可される。一方、部屋20の内側(退室側)には認証装置は設置されておらず、そのため、退室時に認証処理は行われない。認証装置28は、個人識別情報が記憶された記憶媒体から接触又は非接触で個人識別情報を読み取り、読み取った個人識別情報を用いて人を認証し、認証が成功したときに電気錠24に解錠信号を送ることで電気錠24を解錠する。また、認証装置28は、認証結果を入退室管理部12に出力する。個人識別情報が記憶された記憶媒体は、例えばICカードや携帯端末等である。個人識別情報は、部屋20への入室が許可された人の個人情報であり、例えば複数の文字を含むID番号や図形等を含む情報である。認証装置28は、人の指紋や静脈等の生体情報を読み取り、その生体情報を用いて認証を行ってもよい。なお、認証装置28は、解錠信号及び施錠信号を入退室管理部12に出力してもよい。   An authentication device 28 is installed outside the room 20 and in the vicinity of the door 22. As described above, the authentication device 28 is installed on the entrance side of the room 20, and the authentication process is performed when entering the room, and the authenticated person is permitted to enter the room. On the other hand, an authentication device is not installed inside the room 20 (exiting side), and therefore, authentication processing is not performed when leaving the room. The authentication device 28 reads the personal identification information from the storage medium storing the personal identification information in contact or non-contact, authenticates the person using the read personal identification information, and unlocks the electric lock 24 when the authentication is successful. The electric lock 24 is unlocked by sending a lock signal. The authentication device 28 outputs the authentication result to the entrance / exit management unit 12. The storage medium storing the personal identification information is, for example, an IC card or a portable terminal. The personal identification information is personal information of a person who is permitted to enter the room 20, for example, information including an ID number including a plurality of characters, a figure, and the like. The authentication device 28 may read biometric information such as a person's fingerprint or vein and perform authentication using the biometric information. The authentication device 28 may output an unlock signal and a lock signal to the entrance / exit management unit 12.

入退室管理部12は認証装置28から認証結果を受け、その認証結果に基づき、部屋20への人の入室履歴を管理する。入退室管理部12は、入室履歴を示す入室履歴情報を在室人数推定部14に出力する。なお、入退室管理部12は、部屋20への人の入室を検知する検知手段の一例として機能する。   The entrance / exit management unit 12 receives an authentication result from the authentication device 28, and manages a person's entry history into the room 20 based on the authentication result. The room entry / exit management unit 12 outputs room entry history information indicating the room entry history to the occupancy number estimation unit 14. The entrance / exit management unit 12 functions as an example of a detection unit that detects a person entering the room 20.

また、入退室管理部12は、認証装置28によって読み取られた個人識別情報を用いて認証処理を行ってもよい。例えば、入退室管理部12は個人識別情報を記憶し、認証装置28によって読み取られた個人識別情報と、入退室管理部12に記憶されている個人識別情報とを照合することで、部屋20に入室する人を認証する。認証が成功した場合、入退室管理部12は電気錠24に解錠信号を出力する。一方、認証が成功しなかった場合、入退室管理部12は、人の入室を許可せず、電気錠24は施錠状態のままとなる。例えば、認証装置28自体が認証処理を行わずに記録媒体の読取装置として機能する場合、入退室管理部12によって認証処理を行うことが考えられる。この場合、入退室管理部12は、認証装置の一例と機能することになる。   Further, the entrance / exit management unit 12 may perform authentication processing using the personal identification information read by the authentication device 28. For example, the entrance / exit management unit 12 stores the personal identification information, and collates the personal identification information read by the authentication device 28 with the personal identification information stored in the entrance / exit management unit 12, so that the room 20 Authenticate who enters the room. When the authentication is successful, the entrance / exit management unit 12 outputs an unlock signal to the electric lock 24. On the other hand, if the authentication is not successful, the entrance / exit management unit 12 does not permit the person to enter the room, and the electric lock 24 remains locked. For example, when the authentication device 28 itself functions as a reading device for a recording medium without performing authentication processing, it is conceivable that the entrance / exit management unit 12 performs authentication processing. In this case, the entrance / exit management unit 12 functions as an example of an authentication device.

また、入退室管理部12は、扉開閉センサ26から送られる扉開信号及び扉閉信号を用いることで、扉22の開閉履歴を管理する。入退室管理部12は、扉22の開閉履歴を示す扉開閉履歴情報を在室人数推定部14に出力する。例えば、入退室管理部12は、扉開信号を受け付けた時刻(扉開時刻に相当)と、扉閉信号を受け付けた時刻(扉閉時刻に相当)とに基づき、扉22の開閉履歴を管理する。または、扉開信号及び扉閉信号に扉開時刻情報及び扉閉時刻情報が含まれている場合、入退室管理部12は、それらの情報に基づいて扉22の開閉履歴を管理する。なお、解錠信号及び施錠信号が電気錠24又は認証装置28から入退室管理部12に送られてきた場合、入退室管理部12は、解錠信号及び施錠信号を受け付けた時刻に基づいて、扉22の開閉履歴を管理してもよい。例えば、入退室管理部12は、解錠された時刻を扉開時刻とし、施錠された時刻を扉閉時刻として開閉履歴を管理してもよい。   Further, the entrance / exit management unit 12 manages the opening / closing history of the door 22 by using the door opening signal and the door closing signal sent from the door opening / closing sensor 26. The entrance / exit management unit 12 outputs door opening / closing history information indicating the opening / closing history of the door 22 to the occupancy number estimation unit 14. For example, the entrance / exit management unit 12 manages the opening / closing history of the door 22 based on the time when the door open signal is received (corresponding to the door open time) and the time when the door close signal is received (corresponding to the door close time). To do. Alternatively, when the door opening signal and the door closing signal include door opening time information and door closing time information, the entrance / exit management unit 12 manages the opening / closing history of the door 22 based on the information. When the unlocking signal and the locking signal are sent from the electric lock 24 or the authentication device 28 to the entrance / exit management unit 12, the entrance / exit management unit 12 is based on the time when the unlocking signal and the locking signal are received. The opening / closing history of the door 22 may be managed. For example, the entrance / exit management unit 12 may manage the opening / closing history with the unlocked time as the door opening time and the locked time as the door closing time.

また、入退室管理部12は、部屋20に対する警備の状態を管理する。部屋20に対して警備操作が行われると、部屋20への人の入室が禁止され、警備操作が解除されると、部屋20への人の入室が許可される。   Further, the entrance / exit management unit 12 manages the state of security for the room 20. When a security operation is performed on the room 20, a person entering the room 20 is prohibited, and when the security operation is released, a person entering the room 20 is permitted.

ここで、以下の説明で使用する用語(入室人数、退室人数、扉22の開閉時点、扉開時間間隔)について説明する。
入室人数は、部屋20に入室した人数であり、退室人数は、部屋20から退室した人数である。
扉開時間間隔は、扉22が開いてから次に閉じるまでの時間間隔、つまり、1回の開閉(扉開〜扉閉)で扉22が開いている時間の長さである。
扉22の開閉時点は、扉22の1回の開閉(扉開〜扉閉)の期間を示し、以下の説明では、扉22の各開閉時点を単に「各開閉時点」と称する場合がある。
Here, terms used in the following description (number of people entering, number of people leaving, door 22 opening / closing time, door opening time interval) will be described.
The number of people entering the room 20 is the number of people who have entered the room 20, and the number of people leaving the room is the number of people who have left the room 20.
The door opening time interval is a time interval from when the door 22 is opened until it is next closed, that is, a length of time for which the door 22 is opened by one opening and closing (door opening to door closing).
The opening and closing time of the door 22 indicates a period of one opening and closing (door opening to door closing) of the door 22, and in the following description, each opening and closing time of the door 22 may be simply referred to as “each opening and closing time”.

在室人数推定部14は、部屋20への人の入室履歴と扉22の開閉履歴とに基づき、部屋20の在室人数を推定する。例えば、在室人数推定部14は、各開閉時点の扉開時間間隔に基づいて各開閉時点の退室人数を推定し、各開閉時点の退室人数と入室人数とに基づいて各開閉時点の在室人数を求める。具体的には、扉開時間間隔に相当する通行時間と通行人数とに相関があると考え、扉開時間間隔を通行人数に依存した確率変数とし、入室人数に矛盾しない範囲で、事後確率(尤度)が最大となる退室人数を推定する。   The occupancy number estimation unit 14 estimates the occupancy number of the room 20 based on the person's entry history into the room 20 and the door 22 opening / closing history. For example, the occupancy estimation unit 14 estimates the number of people leaving at each opening and closing time based on the door opening time interval at each opening and closing, and based on the number of people leaving and opening at each opening and closing time. Ask for the number of people. Specifically, considering that there is a correlation between the passage time corresponding to the door opening time interval and the number of people passing by, the door opening time interval is a random variable that depends on the number of people passing, and the posterior probability ( Estimate the number of people leaving the room with the highest likelihood.

図2に示すフローチャートを参照して、在室人数推定部14による処理を説明する。一例として、警備解除から警備操作までの期間を、在室人数を求める対象期間とする。例えば、朝に警備が解除されて同日の夜に警備された場合、対象期間は、警備が解除された朝から警備された夜までの期間となる。在室人数推定部14は、対象期間の入室履歴と扉開閉履歴とに基づいて在室人数を推定する。   With reference to the flowchart shown in FIG. 2, the process by the occupancy estimation unit 14 will be described. As an example, the period from the security release to the security operation is a target period for obtaining the number of people in the room. For example, when the security is released in the morning and the night is guarded on the same day, the target period is a period from the morning when the security is released to the night when the security is released. The occupancy estimation unit 14 estimates the occupancy based on the entry history and door opening / closing history of the target period.

まず、在室人数推定部14は、入室履歴及び扉開閉履歴に基づき、1回の扉開閉を1つの履歴グループとしてグループ分けする(S01)。以下、履歴グループ番号をx(=1,2,3,・・・)とする。このとき、扉開直前及び扉開〜扉閉の間の入室履歴もグループに含める。   First, the occupancy estimation unit 14 groups one door opening / closing as one history group based on the room entry history and the door opening / closing history (S01). Hereinafter, the history group number is assumed to be x (= 1, 2, 3,...). At this time, entry histories immediately before the door opening and between the door opening and the door closing are also included in the group.

次に、在室人数推定部14は、入室履歴及び扉開閉履歴に基づき、各履歴グループxについて、扉開時間間隔T(x)、入室人数Nin(x)及び最大在室人数maxS(x)を、以下の式に従って求める(S02)。
扉開時間間隔T(x)=扉閉時刻−扉開時刻
入室人数Nin(x)=履歴グループ内の入室履歴の数
最大在室人数maxS(x)=扉閉時刻以前で同一人物の重複のない入室人数
すなわち、扉開時刻と扉閉時刻とに基づいて扉開時間間隔T(x)を求める。また、入室時には認証処理が行われるため、認証された人数を入室人数として求める。また、以前に認証されて入室した人が再び認証されて入室する場合、2回目以降の入室については、最大在室人数としてカウントしない。
Next, the occupancy estimation unit 14 determines the door opening time interval T (x), the occupancy Nin (x), and the maximum occupancy maxS (x) for each history group x based on the entry history and the door opening / closing history. Is obtained according to the following equation (S02).
Door opening time interval T (x) = door closing time−door opening time number of people entering Nin (x) = number of room entry histories in history group Maximum number of people occupying maxS (x) = duplicate of the same person before door closing time The door opening time interval T (x) is obtained based on the number of people who have not entered, that is, the door opening time and the door closing time. In addition, since authentication processing is performed when entering a room, the number of authenticated persons is obtained as the number of persons entering the room. In addition, when a person who has been authenticated before entering the room is authenticated again, the second and subsequent rooms are not counted as the maximum number of people in the room.

そして、在室人数推定部14は、退室人数Nout(x)の合計が入室人数Nin(x)の合計と一致するように、各履歴グループx(各開閉時点)に任意の退室人数Nout(x)を割り振る(S03)。在室人数推定部14は、退室人数Nout(x)の割り振りが異なる複数の組み合わせ(パターン)を作成する。   Then, the occupancy number estimation unit 14 arbitrarily sets the number of people leaving the room Nout (x) for each history group x (at each opening and closing time) so that the sum of the number of people leaving the room Nout (x) matches the total of the number of people entering the room Nin (x). ) Is allocated (S03). The occupancy number estimation unit 14 creates a plurality of combinations (patterns) with different allocations of the vacant number Nout (x).

次に、在室人数推定部14は、各組み合わせ(各パターン)の各履歴グループxについて、通行人数N(x)及び仮の在室人数S(x)を求める(S04)。通行人数N(x)及び仮の在室人数S(x)は、以下のようにして求められる。なお、推定によって確定される前の在室人数を、確定後の在室人数と区別するために、便宜的に、仮の在室人数S(x)と称することとする。
通行人数N(x)=入室人数Nin(x)+退室人数Nout(x)
すなわち、入室人数Nin(x)と退室人数Nout(x)との和を、通行人数N(x)とする。
また、仮の在室人数を以下の漸化式を用いて求める。
履歴グループ(x=1)の仮の在室人数=Nin(1)−Nout(1)
履歴グループx(x>1)の仮の在室人数=S(x−1)+Nin(x)−Nout(x)
すなわち、前の履歴グループの仮の在室人数と計算対象の履歴グループの入室人数との和から、計算対象の履歴グループの退室人数を減算して得られた値を、計算対象の履歴グループの仮の在室人数とする。また、最初の履歴グループ(x=1)以前には在室者は存在しないため、x=1の開閉時点における入室人数から退室人数を減算して得られた値が、最初の履歴グループ1の在室人数となる。
Next, the occupancy number estimation unit 14 obtains a passing number N (x) and a temporary occupancy number S (x) for each history group x of each combination (each pattern) (S04). The number of people N (x) and the number of people in the room S (x) are obtained as follows. For the sake of convenience, the number of people in the room before being determined by estimation will be referred to as a temporary number of people in the room S (x) in order to distinguish it from the number of people in the room after confirmation.
Number of passengers N (x) = Number of people entering Nin (x) + Number of people leaving the room Nout (x)
That is, the sum of the number of people entering Nin (x) and the number of people leaving Nout (x) is defined as the number of people passing N (x).
Also, the tentative number of people in the room is obtained using the following recurrence formula.
Temporary number of people in history group (x = 1) = Nin (1) −Nout (1)
Temporary number of people in history group x (x> 1) = S (x−1) + Nin (x) −Nout (x)
That is, the value obtained by subtracting the number of people leaving the calculation target history group from the sum of the temporary number of people in the previous history group and the number of people entering the calculation target history group is the value of the calculation target history group. Temporary number of people in the room. Further, since there are no occupants before the first history group (x = 1), the value obtained by subtracting the number of people leaving the room at the time of opening and closing when x = 1 is the value obtained for the first history group 1. Number of people in the room.

在室人数推定部14は、仮の在室人数S(x)が0未満となる組み合わせ(パターン)や、仮の在室人数S(x)が最大在室人数maxS(x)よりも大きくなる組み合わせ(パターン)を破棄する(S05)。すなわち、S(x)<0となる組み合わせや、S(x)>maxS(x)となる組み合わせを破棄する。   The occupancy estimation unit 14 is a combination (pattern) in which the tentative occupancy S (x) is less than 0, or the tentative occupancy S (x) is larger than the maximum occupancy maxS (x). The combination (pattern) is discarded (S05). That is, the combination that satisfies S (x) <0 and the combination that satisfies S (x)> maxS (x) are discarded.

そして、在室人数推定部14は、破棄した組み合わせを除いた各組み合わせ(各パターン)について、扉開時間間隔T(x)と通行人数N(x)とを定式化した通行モデルに基づいて尤度を求め(S06)、尤度が最大となる組み合わせ(パターン)を特定する(S07)。在室人数推定部14は、尤度が最大となる組み合わせ(パターン)の仮の在室人数S(x)を、在室人数の推定値として出力する(S08)。   Then, the occupancy number estimation unit 14 estimates the likelihood based on a passing model in which the door opening time interval T (x) and the passing number N (x) are formulated for each combination (each pattern) excluding the discarded combination. The degree is obtained (S06), and the combination (pattern) that maximizes the likelihood is specified (S07). The occupancy number estimation unit 14 outputs the tentative occupancy number S (x) of the combination (pattern) with the maximum likelihood as an estimated value of the occupancy number (S08).

以下、尤度を求める方法について説明する。通行モデルは、一例として以下の式で表される。
通行モデル:扉開時間間隔T(x)=α×通行人数N(x)+β+ε
α、β及びεは、以下の通りである。
α=1人が扉22の通行に要する時間の長さ
β=扉22が開くのに要する時間の長さ+扉22が閉じるのに要する時間の長さ
ε=誤差項(正規分布で仮定)、通行間隔
Hereinafter, a method for obtaining the likelihood will be described. The traffic model is represented by the following formula as an example.
Traffic model: Door opening time interval T (x) = α × passenger number N (x) + β + ε
α, β, and ε are as follows.
α = length of time required for one person to pass through the door 22 β = length of time required for the door 22 to open + length of time required for the door 22 to close ε = error term (assumed by normal distribution) , Traffic interval

在室人数推定部14は、確率密度Pdf(x)から尤度(尤度の比例値)を求める。正規分布を仮定し、各履歴グループxの確率密度Pdf(x)を以下の式(1)から求める。ここで、σはεの分散である。なお、各グループの確率密度Pdf(x)は、各グループの尤度ということができる。

Figure 0005875541
The occupancy estimation unit 14 obtains likelihood (a likelihood proportional value) from the probability density Pdf (x). Assuming a normal distribution, the probability density Pdf (x) of each history group x is obtained from the following equation (1). Here, σ is the variance of ε. Note that the probability density Pdf (x) of each group can be referred to as the likelihood of each group.
Figure 0005875541

そして、在室人数推定部14は、以下の式に示すように確率密度Pdf(x)の積を求め、これを尤度の比例値とする。以下では、尤度の比例値を尤度として説明する。
尤度∝Pdf(1)×Pdf(2)×・・・×Pdf(x)
在室人数推定部14は、各組み合わせ(パターン)について尤度を求め、尤度が最大となる組み合わせ(パターン)の仮の在室人数S(x)を、在室人数の推定値として出力する。
Then, the occupancy estimation unit 14 obtains a product of the probability density Pdf (x) as shown in the following formula, and sets this as a proportional value of likelihood. Below, the proportional value of likelihood is demonstrated as likelihood.
Likelihood ∝ Pdf (1) x Pdf (2) x ... x Pdf (x)
The occupancy estimation unit 14 obtains the likelihood for each combination (pattern), and outputs the tentative occupancy S (x) of the combination (pattern) with the maximum likelihood as an estimated value of the occupancy. .

出力部16は、在室人数推定部14によって推定された部屋20の在室人数を出力する。出力部16は、一例として、在室人数を表示する表示装置、在室人数を示す情報を外部に送信するネットワークインターフェース又は印刷装置等である。   The output unit 16 outputs the occupancy number of the room 20 estimated by the occupancy number estimation unit 14. For example, the output unit 16 is a display device that displays the number of people in the room, a network interface that transmits information indicating the number of people in the room, a printing device, or the like.

(動作例)
次に、図3及び図4を参照して、入退室管理装置10の具体的な動作例を説明する。
(Operation example)
Next, with reference to FIG.3 and FIG.4, the specific operation example of the entrance / exit management apparatus 10 is demonstrated.

例えば図3に示すように、時刻8:45:03の時点で部屋20の警備が解除されてA氏が部屋20に入室し、時刻9:00:04〜9:00:08の間で扉22が開閉し、時刻9:30:05の時点でB氏が部屋20に入室し、時刻13:45:23の時点でA氏が部屋20に入室し、時刻17:00:01〜17:00:11の間で扉22が開閉している。A,B氏は認証装置28によって認証されることで入室が許可され、個人識別情報や認証結果が入退室管理部12に入力される。これにより、入退室管理部12は、A,B氏の入室を検知する。また、入退室管理部12は、扉開信号及び扉閉信号(又は、解錠信号及び施錠信号)に基づいて扉22の開閉を検知する。   For example, as shown in FIG. 3, at time 8:45:03, the security of the room 20 is released and Mr. A enters the room 20, and the door is between 9:00:04 and 9:00:08. 22 opens and closes, Mr. B enters the room 20 at time 9:30:05, Mr. A enters the room 20 at time 13:45:23, and 17: 00: 01-17: The door 22 opens and closes between 00:11. Mr. A and B are permitted to enter the room by being authenticated by the authentication device 28, and personal identification information and an authentication result are input to the entrance / exit management unit 12. Thereby, the entrance / exit management unit 12 detects the entrance of Mr. A and B. Moreover, the entrance / exit management part 12 detects opening / closing of the door 22 based on a door open signal and a door close signal (or an unlock signal and a lock signal).

以上のような動作例の場合、入退室管理装置10は以下のようにして在室人数を推定する。まず、入退室管理部12は、図3に示す入室履歴情報及び扉開閉履歴情報を在室人数推定部14に出力する。在室人数推定部14は、1回の扉開閉を1つの履歴グループとしてグループ分けする。グループ分けの一例を、図4(a)のテーブルに示す。具体的には、以下のように履歴グループ1〜5にグループ分けする。
履歴グループ1:A氏が入室した時刻8:45:03〜扉閉時刻8:45:07
履歴グループ2:扉開時刻9:00:04〜扉閉時刻9:00:08
履歴グループ3:B氏が入室した時刻9:30:05〜扉閉時刻9:30:10
履歴グループ4:A氏が入室した時刻13:45:23〜扉閉時刻13:45:26
履歴グループ5:扉開時刻17:00:01〜扉閉時刻17:00:11
In the case of the above operation example, the entrance / exit management device 10 estimates the number of people in the room as follows. First, the entrance / exit management unit 12 outputs the room entry history information and door opening / closing history information shown in FIG. The occupancy estimation unit 14 groups one door opening / closing as one history group. An example of grouping is shown in the table of FIG. Specifically, the history groups 1 to 5 are grouped as follows.
History group 1: Time when Mr. A entered the room 8:45:03 to door closing time 8:45:07
History group 2: Door opening time 9:00:04 to door closing time 9:00:08
History group 3: Time when Mr. B entered the room 9:30:05 to door close time 9:30:10
History group 4: Time when Mr. A entered the room 13:45:23 to door closing time 13:45:26
History group 5: Door opening time 17:00:01 to door closing time 17:00:11

そして、在室人数推定部14は、履歴グループ1〜5について、扉開時間間隔T(x)、入室人数Nin(x)及び最大在室人数maxS(x)を求める。各値は、図4(a)のテーブルに示すように次のようになる。   Then, the occupancy estimation unit 14 obtains the door opening time interval T (x), the occupancy Nin (x), and the maximum occupancy maxS (x) for the history groups 1 to 5. Each value is as follows as shown in the table of FIG.

履歴グループ1:
扉開時間間隔T(1)=4、入室人数Nin(1)=1、最大在室人数maxS(1)=1
A氏1人の入室が検知されているため、上記の値となる。
History group 1:
Door opening time interval T (1) = 4, number of people entering Nin (1) = 1, maximum number of people occupying maxS (1) = 1
Since the entry of Mr. A is detected, the above value is obtained.

履歴グループ2:
扉開時間間隔T(2)=4、入室人数Nin(2)=0、最大在室人数maxS(2)=1
History group 2:
Door opening time interval T (2) = 4, number of people entering Nin (2) = 0, maximum number of people occupying maxS (2) = 1

履歴グループ3:
扉開時間間隔T(3)=5、入室人数Nin(3)=1、最大在室人数maxS(3)=2
B氏1人の入室が検知されているため、入室人数Nin(3)=1となり、認証の結果、A氏とB氏は同一人物ではなく重複しないため、最大在室人数maxS(3)=2となる。
History group 3:
Door opening time interval T (3) = 5, number of people entering Nin (3) = 1, maximum number of people occupying maxS (3) = 2
Since the entry of Mr. B is detected, the number of people entering the room is Nin (3) = 1, and as a result of authentication, Mr. A and Mr. B are not the same person and do not overlap, so the maximum number of people in the room maxS (3) = 2

履歴グループ4:
扉開時間間隔T(4)=3、入室人数Nin(4)=1、最大在室人数maxS(4)=2
A氏1人の入室が検知されているため、入室人数Nin(4)=1となり、認証の結果、A氏の入室は2回目であり重複するため、最大在室人数maxS(4)=2のままである。
History group 4:
Door opening time interval T (4) = 3, number of people entering Nin (4) = 1, maximum number of people occupying maxS (4) = 2
Since the entry of Mr. A is detected, the number of people entering the room is Nin (4) = 1. As a result of the authentication, Mr. A's entry is the second time and overlaps, so the maximum number of people in the room maxS (4) = 2. Remains.

履歴グループ5:
扉開時間間隔T(5)=10、入室人数Nin(5)=0、最大在室人数maxS(5)=0
History group 5:
Door opening time interval T (5) = 10, number of people entering Nin (5) = 0, maximum number of people occupying maxS (5) = 0

次に、在室人数推定部14は、退室人数Nout(x)の合計が入室人数Nin(x)の合計と一致するように、履歴グループ1〜5に任意の退室人数Nout(x)を割り振る。さらに、割り振りが異なる複数の組み合わせ(パターン)を作成する。この動作例では、入室人数Nin(x)の合計は3人であるため、退室人数Nout(x)の合計が3人となるように、退室人数Nout(x)を履歴グループ1〜5に割り振る。例えば図4(b)に示すように、パターン1(組み合わせ1)における退室人数Nout(x)は、次のように割り振られている。
Nout(1)=0
Nout(2)=1
Nout(3)=0
Nout(4)=0
Nout(5)=2
Next, the occupancy estimation unit 14 assigns an arbitrary vacancy number Nout (x) to the history groups 1 to 5 so that the total of the vacancy number Nout (x) matches the total of the occupancy number Nin (x). . Furthermore, a plurality of combinations (patterns) with different allocations are created. In this operation example, since the total number of entering people Nin (x) is 3, the leaving number Nout (x) is allocated to the history groups 1 to 5 so that the total number of leaving people Nout (x) is 3. . For example, as shown in FIG. 4B, the number of people leaving the room Nout (x) in pattern 1 (combination 1) is allocated as follows.
Nout (1) = 0
Nout (2) = 1
Nout (3) = 0
Nout (4) = 0
Nout (5) = 2

また、図4(c)に示すように、パターン2(組み合わせ2)における退室人数Nout(x)は、次のように割り振られている。
Nout(1)=0
Nout(2)=0
Nout(3)=0
Nout(4)=0
Nout(5)=3
Moreover, as shown in FIG.4 (c), the leaving number Nout (x) in the pattern 2 (combination 2) is allocated as follows.
Nout (1) = 0
Nout (2) = 0
Nout (3) = 0
Nout (4) = 0
Nout (5) = 3

退室人数Nout(x)の割り振り方によって、パターン1,2以外のパターンが作成されるが、説明の都合上、パターン1,2のみを図示して説明する。   Patterns other than patterns 1 and 2 are created depending on how the number of people leaving the room Nout (x) is allocated. For convenience of explanation, only patterns 1 and 2 are illustrated and described.

そして、在室人数推定部14は、各パターン(各組み合わせ)の履歴グループ1〜5について、通行人数N(x)及び仮の在室人数S(x)を求める。例えば図4(b)に示すように、パターン1(組み合わせ1)における通行人数N(x)及び仮の在室人数S(x)は、次のようになる。   Then, the occupancy number estimation unit 14 obtains the passing number N (x) and the temporary occupancy number S (x) for the history groups 1 to 5 of each pattern (each combination). For example, as shown in FIG. 4B, the number of people N (x) and the number of people in the room S (x) in the pattern 1 (combination 1) are as follows.

履歴グループ1:
通行人数N(1)=1(Nin(1)+Nout(1)=1)
仮の在室人数S(1)=1(Nin(1)−Nout(1)=1)
History group 1:
Number of passengers N (1) = 1 (Nin (1) + Nout (1) = 1)
Temporary number of people S (1) = 1 (Nin (1) −Nout (1) = 1)

履歴グループ2:
通行人数N(2)=1(Nin(2)+Nout(2)=1)
仮の在室人数S(2)=0(S(1)+Nin(2)−Nout(2)=0)
History group 2:
Number of passengers N (2) = 1 (Nin (2) + Nout (2) = 1)
Temporary number of people S (2) = 0 (S (1) + Nin (2) −Nout (2) = 0)

履歴グループ3:
通行人数N(3)=1(Nin(3)+Nout(3)=1)
仮の在室人数S(3)=1(S(2)+Nin(3)−Nout(3)=1)
History group 3:
Number of passengers N (3) = 1 (Nin (3) + Nout (3) = 1)
Temporary number of people S (3) = 1 (S (2) + Nin (3) −Nout (3) = 1)

履歴グループ4:
通行人数N(4)=1(Nin(4)+Nout(4)=1)
仮の在室人数S(4)=2(S(3)+Nin(4)−Nout(4)=2)
History group 4:
Number of passengers N (4) = 1 (Nin (4) + Nout (4) = 1)
Temporary number of people S (4) = 2 (S (3) + Nin (4) −Nout (4) = 2)

履歴グループ5:
通行人数N(5)=2(Nin(5)+Nout(5)=2)
仮の在室人数S(5)=0(S(4)+Nin(5)−Nout(5)=0)
History group 5:
Number of passengers N (5) = 2 (Nin (5) + Nout (5) = 2)
Temporary number of people S (5) = 0 (S (4) + Nin (5) −Nout (5) = 0)

また、図4(c)に示すように、パターン2(組み合わせ2)における通行人数N(x)及び仮の在室人数S(x)は、次のようになる。
履歴グループ1:通行人数N(1)=1、仮の在室人数S(1)=1
履歴グループ2:通行人数N(2)=0、仮の在室人数S(2)=1
履歴グループ3:通行人数N(3)=1、仮の在室人数S(3)=2
履歴グループ4:通行人数N(4)=1、仮の在室人数S(4)=3
履歴グループ5:通行人数N(5)=3、仮の在室人数S(5)=0
Further, as shown in FIG. 4C, the number of people N (x) and the number of people in the room S (x) in the pattern 2 (combination 2) are as follows.
History group 1: Passage number N (1) = 1, Temporary occupancy number S (1) = 1
History group 2: Passage number N (2) = 0, Temporary occupancy number S (2) = 1
History group 3: Passage number N (3) = 1, Temporary occupancy number S (3) = 2
History group 4: Passenger number N (4) = 1, Temporary occupant number S (4) = 3
History group 5: Passage number N (5) = 3, Temporary occupancy number S (5) = 0

在室人数推定部14は、仮の在室人数S(x)が0未満となるパターン(組み合わせ)や、仮の在室人数S(x)が最大在室人数maxS(x)よりも大きくなるパターンを破棄する。図4(c)に示すパターン2の履歴グループ4に注目すると、最大在室人数maxS(4)=2であり、在室人数S(4)=3であるため、S(4)>maxS(4)となる。従って、在室人数推定部14は、パターン2を破棄し、パターン2については尤度を求めないことにする。   The occupancy estimation unit 14 has a pattern (combination) in which the tentative occupancy S (x) is less than 0, or the tentative occupancy S (x) is larger than the maximum occupancy maxS (x). Discard the pattern. When attention is paid to the history group 4 of the pattern 2 shown in FIG. 4C, since the maximum occupancy maxS (4) = 2 and the occupancy S (4) = 3, S (4)> maxS ( 4). Therefore, the occupancy estimation unit 14 discards the pattern 2 and does not obtain the likelihood for the pattern 2.

そして、在室人数推定部14は、破棄したパターン(組み合わせ)を除いた各パターンについて、上記式(1)に従って確率密度及び尤度を求め、尤度が最大となるパターンの仮の在室人数S(x)を、在室人数の推定値として出力する。例えば、パターン1(組み合わせ1)の尤度が最大となった場合、在室人数推定部14は、各履歴グループ(各開閉時点)における仮の在室人数S(x)を、在室人数の推定値として出力する。   Then, the occupancy estimation unit 14 obtains the probability density and the likelihood according to the above equation (1) for each pattern excluding the discarded pattern (combination), and the tentative occupancy in the pattern with the maximum likelihood. S (x) is output as an estimated value of the number of people in the room. For example, when the likelihood of pattern 1 (combination 1) is maximized, the occupancy estimation unit 14 calculates the tentative occupancy S (x) in each history group (at each opening and closing time) as the occupancy occupancy. Output as an estimated value.

以上のように、部屋20の入室側(外側)に設置された認証装置28を用いて人の入室を検知し、扉開時間間隔を用いて退室人数を推定することで、部屋20の退室側(内側)に認証装置を設けなくても、部屋20の在室人数を推定することが可能となる。また、通行人数をカウントするセンサやカメラ等を追加で設置する必要もない。従って、部屋の入室側に認証装置が設置されている場合、コストを増大させずに当該部屋の在室人数を推定することが可能となる。   As described above, the entrance side of the room 20 is detected by detecting the entrance of the person using the authentication device 28 installed on the entrance side (outside) of the room 20 and estimating the number of persons leaving the room using the door opening time interval. The number of people in the room 20 can be estimated without providing an authentication device (inside). In addition, there is no need to additionally install a sensor or camera for counting the number of people passing by. Therefore, when the authentication device is installed on the entrance side of the room, the number of people in the room can be estimated without increasing the cost.

次に、図5のグラフを参照して本実施形態に係る入退室管理装置10の効果を説明する。図5において、横軸は日付であり、縦軸は在室人数である。グラフ100(黒のプロット)は、本実施形態による処理によって推定された在室人数を示しており、部屋の入室側(外側)のみに認証装置を設置した条件で推定された人数である。一方、グラフ200(白のプロット)は、部屋の入室側(外側)と退室側(内側)とに認証装置を設置した条件で求められた在室人数である。具体的には、入室側に設置された認証装置にて入室人数を検知し、退室側に設置された認証装置にて退室人数を検知し、入室人数と退室人数とから在室人数を求めた結果がグラフ200である。グラフ100,200はともに、一例として、人が部屋に30分間在室した場合を0.5人、60分間在室した場合を1.0人として、在室時間の割合を積算した結果である。   Next, the effect of the entrance / exit management device 10 according to the present embodiment will be described with reference to the graph of FIG. In FIG. 5, the horizontal axis is the date, and the vertical axis is the number of people in the room. A graph 100 (black plot) shows the number of people in the room estimated by the processing according to the present embodiment, and is the number of people estimated under the condition that the authentication device is installed only on the room entrance side (outside). On the other hand, the graph 200 (white plot) is the number of people in the room determined under the condition that the authentication device is installed on the entrance side (outside) and exit side (inside) of the room. Specifically, the number of people entering the room was detected by the authentication device installed on the entrance side, the number of people leaving the room was detected by the authentication device installed on the exit side, and the number of people in the room was calculated from the number of people entering and leaving the room. The result is a graph 200. Both graphs 100 and 200 are, for example, the results of accumulating the ratio of the occupancy time, assuming that 0.5 person is in the room for 30 minutes and 1.0 person is in the room for 60 minutes. .

グラフ100,200を比較すると、在室人数の増減の傾向はほぼ一致していると評価することができる。すなわち、本実施形態による推定処理によると、部屋の入室側(外側)のみに認証装置が設置されている場合であっても、正確に在室人数を推定することが可能となる。換言すると、入室側(外側)のみに認証装置が設置されている部屋に対して、退室側(内側)に認証装置を追加で設置しなくても、在室人数を正確に推定することができる。また、カメラやセンサ等を追加する必要もない。このように、認証装置やカメラ等を追加する必要がないため、コストを増大させずに在室人数をより正確に推定することが可能となる。   Comparing the graphs 100 and 200, it can be evaluated that the tendency of increase / decrease in the number of people in the room is almost the same. That is, according to the estimation process according to the present embodiment, the number of people in the room can be accurately estimated even when the authentication device is installed only on the entrance side (outside) of the room. In other words, the number of people in the room can be accurately estimated without installing an additional authentication device on the exit side (inside) with respect to a room where the authentication device is installed only on the entrance side (outside). . Further, it is not necessary to add a camera, a sensor, or the like. Thus, since there is no need to add an authentication device or a camera, the number of people in the room can be estimated more accurately without increasing the cost.

(変形例1)
次に、図6に示すフローチャート及び図7を参照して変形例1を説明する。上述した実施形態では、任意の退室人数Nout(x)を各履歴グループに割り振ることで複数の組み合わせを作成する場合について説明した。変形例1では、尤度(確率密度Pdf(x))が最大となるように退室人数Nout(x)を1人ずつ割り振っていくことで、計算の高速化を図る。すなわち、在室人数推定部14は、各履歴グループの退室人数Nout(x)を所定値として通行人数N(x)を求め、扉開時間間隔T(x)と通行人数N(x)とに基づいて各履歴グループの確率密度Pdf(x)を求め、確率密度Pdf(x)が最小となる履歴グループの仮の退室人数Nout(x)を変更し、尤度を更新する。そして、在室人数推定部14は、仮の在室人数S(x)>最大在室人数maxS(x)となる履歴グループが無くなるまで、退室人数Nout(x)の変更及び尤度の更新を繰り返し実行し、最終的に得られた仮の在室人数S(x)を、在室人数の推定値として出力する。以下、具体例を挙げて説明する。
(Modification 1)
Next, Modification 1 will be described with reference to the flowchart shown in FIG. 6 and FIG. In the above-described embodiment, a case has been described in which a plurality of combinations are created by assigning an arbitrary number of people leaving the room Nout (x) to each history group. In the first modification, the number of people leaving the room Nout (x) is allocated one by one so that the likelihood (probability density Pdf (x)) is maximized, thereby speeding up the calculation. That is, the occupancy number estimating unit 14 obtains the number of people passing by N (x) using the number of people leaving the history group Nout (x) as a predetermined value, and sets the door opening time interval T (x) and the number of people passing by N (x). Based on this, the probability density Pdf (x) of each history group is obtained, the temporary leaving number Nout (x) of the history group that minimizes the probability density Pdf (x) is changed, and the likelihood is updated. Then, the occupancy number estimation unit 14 changes the vacant number Nout (x) and updates the likelihood until there is no history group in which the tentative occupancy number S (x)> the maximum occupancy number maxS (x). The temporary occupancy number S (x) finally obtained is repeatedly output as an estimated value of the occupancy number. Hereinafter, a specific example will be described.

まず、在室人数推定部14は、図3に示す入退室履歴情報及び扉開閉履歴情報を受けると、入室履歴及び扉開閉履歴に基づき、図4(a)のテーブルに示すように、1回の扉開閉を1つの履歴グループとしてグループ分けする(S10)。そして、在室人数推定部14は、図2に示すステップS02と同様に、入室履歴及び扉開閉履歴に基づき、各履歴グループxについて、扉開時間間隔T(x)、入室人数Nin(x)及び最大在室人数maxS(x)を求める。(S11)。   First, when the occupancy estimation unit 14 receives the entry / exit history information and the door opening / closing history information shown in FIG. 3, based on the entry history and the door opening / closing history, as shown in the table of FIG. Are grouped as one history group (S10). Then, similarly to step S02 shown in FIG. 2, the occupancy number estimation unit 14 determines the door opening time interval T (x) and the occupancy number Nin (x) for each history group x based on the room entry history and the door opening / closing history. And the maximum number of people in the room maxS (x). (S11).

次に、図7(a)に示すように、在室人数推定部14は、すべての履歴グループの退室人数Nout(x)を所定値(例えば0人)とし(S12)、通行人数N(x)、仮の在室人数S(x)、確率密度Pdf(x)及び尤度を計算する(S13)。   Next, as shown in FIG. 7A, the occupancy number estimation unit 14 sets the number of people leaving the room Nout (x) of all the history groups to a predetermined value (for example, 0 person) (S12), and the number of people N (x ), Tentative occupancy S (x), probability density Pdf (x), and likelihood are calculated (S13).

そして、在室人数推定部14は、履歴グループの番号が小さい方(時刻が相対的に早い履歴グループ)から検索し、最初に仮の在室人数S(x)=0となる履歴グループに、(x_sub_begin)を設定する(S14)。仮の在室人数S(x)=0となる履歴グループが存在しない場合、最初の履歴グループ1に(x_sub_begin)を設定する(x_sub_begin=1)。図7(a)に示す例では、仮の在室人数S(x)=0となる履歴グループが存在しないため、(x_sub_begin=1)とする。   Then, the occupancy estimation unit 14 searches from a history group with a smaller history group number (history group with a relatively early time), and first sets the history group in which the tentative occupancy S (x) = 0. (X_sub_begin) is set (S14). When there is no history group in which the tentative number of people S (x) = 0, (x_sub_begin) is set to the first history group 1 (x_sub_begin = 1). In the example shown in FIG. 7A, since there is no history group in which the tentative number of people S (x) = 0, (x_sub_begin = 1) is set.

次に、在室人数推定部14は、履歴グループの番号が小さい方(時間が相対的に早い履歴グループ)から検索し、最初に仮の在室人数S(x)>最大在室人数maxS(x)となる履歴グループに、(x_sub_end)を設定する(S15)。図7(a)に示すように、履歴グループ4の仮の在室人数S(4)は3人であり、図4(a)に示すように、履歴グループ4の最大在室人数maxS(4)は2人であり、仮の在室人数S(4)>最大在室人数maxS(4)となるため、履歴グループ4に(x_sub_end)を設定する(x_sub_end=4)。   Next, the occupancy estimation unit 14 searches from the one with the smaller history group number (history group with relatively earlier time), and first tentative occupancy S (x)> maximum occupancy maxS ( (x_sub_end) is set to the history group that becomes x) (S15). As shown in FIG. 7A, the tentative occupancy S (4) of the history group 4 is 3, and the maximum occupancy maxS (4 of the history group 4 is 4 as shown in FIG. 4A. ) Is two persons, and the tentative occupancy number S (4)> maximum occupant number maxS (4), (x_sub_end) is set in the history group 4 (x_sub_end = 4).

そして、(x_sub_begin)が設定された履歴グループと(x_sub_end)が設定された履歴グループとの間にある履歴グループのうち、退室人数を1人割り振った場合に尤度の増分が最大となる履歴グループに、退室人数を1人割り振る(S16)。すなわち、
「Nout(x)に1を加算したときのPdf(x)−現在のPdf(x)」
が最大となる履歴グループxに退室人数を1人割り振る。
退室人数を割り振った場合に尤度の増分が最大となる履歴グループにおいては、1人割り振る前の退室人数(0人)となる可能性が他の履歴グループよりも低いため、退室人数を1人割り振ることとする。図7(a)に示す例では、履歴グループ1と履歴グループ4との間の履歴グループのうち、退室人数を1人割り振った場合に尤度の増分が最大となるのは履歴グループ2であるため、履歴グループ2の退室人数Nout(2)に1人を割り振る。この割り振りに伴い、在室人数推定部14は、仮の在室人数S(x)と尤度とを更新する(S17)。図7(b)に、更新後の値を示す。
Among history groups between the history group in which (x_sub_begin) is set and the history group in which (x_sub_end) is set, the history group that maximizes the likelihood when the number of people leaving the room is allocated. The number of people leaving the room is allocated to one person (S16). That is,
“Pdf (x) when 1 is added to Nout (x) −current Pdf (x)”
Allocate one person to the history group x where the maximum is.
In the history group where the likelihood increase is maximum when the number of people leaving the room is allocated, the possibility of becoming the number of people leaving the room (0 people) before allocating one person is lower than that in other history groups. It will be allocated. In the example shown in FIG. 7A, among the history groups between the history group 1 and the history group 4, the history group 2 has the largest increase in likelihood when one person leaving the room is allocated. Therefore, one person is allocated to the number of people leaving the history group 2 Nout (2). Along with this allocation, the occupancy estimation unit 14 updates the tentative occupancy S (x) and the likelihood (S17). FIG. 7B shows the updated values.

そして、仮の在室人数S(x)>最大在室人数maxS(x)となる履歴グループが存在する場合(S18,Yes)、上記のステップS14〜S17の処理を繰り返す。図7(b)に示すように、履歴グループ5の仮の在室人数S(5)は2人であり、図4(a)に示すように、履歴グループ5の最大在室人数maxS(5)は0人であるため、履歴グループ5において、仮の在室人数S(5)>最大在室人数maxS(5)となっている。従って、上記のステップS14〜S17の処理を繰り返す。   If there is a history group such that the tentative occupancy number S (x)> maximum occupancy number maxS (x) (S18, Yes), the processes in steps S14 to S17 are repeated. As shown in FIG. 7B, the temporary occupancy number S (5) of the history group 5 is 2, and the maximum occupancy number maxS (5 of the history group 5 is shown in FIG. 4A. ) Is 0, in the history group 5, the tentative occupancy number S (5)> the maximum occupancy number maxS (5). Therefore, the processes in steps S14 to S17 are repeated.

図7(b)に示す例では、履歴グループ2の仮の在室人数S(2)が0人であるため、図7(c)に示すように、履歴グループ2に、(x_sub_begin)を設定する(x_sub_begin=2)。また、仮の在室人数S(5)>最大在室人数maxS(5)となっているため、履歴グループ5に(x_sub_end)を設定する(x_sub_end=5)。そして、ステップS16,S17の処理を実行することで尤度を更新する。   In the example shown in FIG. 7B, since the temporary occupancy S (2) of the history group 2 is 0, as shown in FIG. 7C, (x_sub_begin) is set in the history group 2. (X_sub_begin = 2). Further, since the tentative occupancy number S (5)> maximum occupancy number maxS (5), (x_sub_end) is set in the history group 5 (x_sub_end = 5). And likelihood is updated by performing the process of step S16, S17.

一方、仮の在室人数S(x)>最大在室人数maxS(x)となる履歴グループが無くなった場合(S18,No)、在室人数推定部14は、最終的に得られた仮の在室人数S(x)を、在室人数の推定値として出力する(S19)。   On the other hand, when there is no history group satisfying the tentative occupancy number S (x)> the maximum occupancy number maxS (x) (No in S18), the occupancy number estimation unit 14 finally obtains the temporary tentative number obtained. The occupant number S (x) is output as an estimated value of the occupant number (S19).

以上のように、尤度(確率密度Pdf(x))が最大となるように退室人数を1人ずつ割り振ることで、退室人数を任意に割り振る場合と比べて、より高速に在室人数を推定することが可能となる。すなわち、複数の組み合わせ(パターン)について尤度等を計算する必要がないため、その分、高速に在室人数を推定することが可能となる。   As described above, by assigning the number of people leaving the room one by one so that the likelihood (probability density Pdf (x)) is maximized, the number of people in the room can be estimated at a higher speed than when the number of people leaving the room is arbitrarily assigned. It becomes possible to do. That is, since it is not necessary to calculate the likelihood or the like for a plurality of combinations (patterns), it is possible to estimate the number of people in the room at a high speed.

(変形例2)
次に、変形例2について説明する。上記の実施形態及び変形例1では、認証処理によって入室人数を把握することを前提としているため、共連れで部屋20に入室した場合には、在室人数の推定に誤差が生じることがある。そこで、共連れ入室があることを考慮し、確率値が最大となるように、入室人数及び退室人数をそれぞれ追加していくことで、共連れ等によって入室人数の一部を把握できない場合であっても、在室人数の検知の精度を向上させる。
(Modification 2)
Next, Modification 2 will be described. In the above-described embodiment and Modification 1, it is assumed that the number of people entering the room is ascertained by the authentication process. Therefore, when entering the room 20 together, an error may occur in the estimation of the number of people in the room. Therefore, considering that there are co-entrance rooms, the number of people entering the room and the number of people leaving the room are added so that the probability value is maximized. Even improve the accuracy of detecting the number of people in the room.

まず、在室人数推定部14は、上記の実施形態及び変形例1で説明した処理によって、退室人数Nout(x)、通行人数N(x)、仮の在室人数S(x)、確率密度Pdf(x)及び尤度を求める(S20)。   First, the occupancy number estimation unit 14 performs the processing described in the above embodiment and the first modification, so that the number of people leaving the room Nout (x), the number of people passing by N (x), the number of people occupying the room S (x), and the probability density Pdf (x) and likelihood are obtained (S20).

次に、在室人数推定部14は、任意の履歴グループの入室人数Nin(x)を1人増加させ、また、仮の在室人数S(x)が0人未満とならないように、任意の履歴グループの退室人数Nout(x)を1人増加させる(S21)。   Next, the occupancy number estimation unit 14 increases the occupancy number Nin (x) of an arbitrary history group by one, and the tentative occupancy number S (x) does not become less than 0. The number of people leaving the history group Nout (x) is increased by one (S21).

そして、在室人数推定部14は尤度を再計算し、尤度が増加した場合、そのときの値(入室人数Nin(x)、退室人数Nout(x)、仮の在室人数S(x))を採用し、尤度が増加しなかった場合、ステップS21で増加させた入室人数Nin(x)及び退室人数Nout(x)を元の値に戻す(S22)。   Then, the occupancy number estimation unit 14 recalculates the likelihood, and when the likelihood increases, the values at that time (number of people entering Nin (x), number of people leaving Nout (x), temporary occupant number S (x )) And the likelihood does not increase, the number Nin (x) of entering rooms and the number Nout (x) of leaving rooms increased in step S21 are returned to the original values (S22).

そして、在室人数推定部14は、尤度が増加しなくなるまでステップS21,S22の処理を繰り返し、尤度が増加しなくなったときの仮の在室人数S(x)を、在室人数の推定値として出力する(S23)。   Then, the occupancy estimation unit 14 repeats the processes of steps S21 and S22 until the likelihood does not increase, and the provisional occupancy S (x) when the likelihood does not increase is obtained as the occupancy occupancy. An estimated value is output (S23).

変形例2を上記の実施形態に適用した場合、在室人数推定部14は、各組み合わせ(パターン)についてステップS21〜S23の処理を実行し、尤度が最大となる組み合わせ(パターン)の仮の在室人数S(x)を、在室人数の推定値として出力する。一方、変形例2を上記の変形例1に適用した場合、仮の在室人数S(x)>最大在室人数maxS(x)となる履歴グループが無くなったときの各履歴グループを対象にして、ステップS21〜S23の処理を実行することで、在室人数等を修正する。   When the modified example 2 is applied to the above-described embodiment, the occupancy number estimation unit 14 performs the processing of steps S21 to S23 for each combination (pattern), and the temporary combination of the combination (pattern) that maximizes the likelihood. The occupant number S (x) is output as an estimated value of the occupant number. On the other hand, when the second modification is applied to the first modification, the history groups when the tentative occupancy number S (x)> the maximum occupancy number maxS (x) no longer exist are targeted. The number of people in the room is corrected by executing the processes of steps S21 to S23.

入室人数Nin(x)及び退室人数Nout(x)を増加させた場合に尤度が増加するということは、認証処理によって検知されなかった入室者が存在する可能性がある。従って、変形例2のように、入室人数Nin(x)及び退室人数Nout(x)を調整して尤度の変化を確認することで、共連れ入室の可能性を検知し、在室人数を修正することが可能となる。その結果、在室人数をより正確に推定することが可能となる。   When the number of people entering Nin (x) and the number of people leaving Nout (x) are increased, the likelihood increases that there may be people who have not been detected by the authentication process. Therefore, as in the second modification, by adjusting the number of people entering Nin (x) and the number of people leaving Nout (x) and confirming the change in likelihood, the possibility of shared entry is detected, and the number of people in the room is determined. It becomes possible to correct. As a result, the number of people in the room can be estimated more accurately.

(変形例3)
次に、変形例3を説明する。変形例3では、通行モデルで用いたパラメータα,β,σを、推定された通行人数から回帰分析によって推定し、推定されたパラメータα,β,σを用いて、再び通行人数を推定する。そして、予め設定された条件を満たすまで上記の推定を繰り返すことで、パラメータα,β,σを決定し、決定したパラメータα,β,σを用いて通行人数を推定する。
(Modification 3)
Next, Modification 3 will be described. In Modification 3, the parameters α, β, and σ used in the traffic model are estimated by regression analysis from the estimated number of traffic, and the number of traffic is estimated again using the estimated parameters α, β, and σ. Then, by repeating the above estimation until a preset condition is satisfied, the parameters α, β, and σ are determined, and the number of passing persons is estimated using the determined parameters α, β, and σ.

具体的には、EMアルゴリズムを用いることでパラメータα,β,σを決定する。まず、在室人数推定部14は、予め決定されたパラメータα,β,σの暫定値を用い、上述したように、通行モデルによって通行人数を推定する(Eステップ)。次に、在室人数推定部14は、Eステップで推定された通行人数を用い、回帰分析(最小二乗法)によってパラメータα,β,σを求める(Mステップ)。そして、Mステップで求められたパラメータα,β,σを新たな暫定値として通行モデルによって通行人数を推定し(Eステップ)、その推定された通行人数を用いて回帰分析によってパラメータα,β,σを求める(Mステップ)。このように、EステップとMステップとを繰り返し実行し、パラメータα,β,σの変化量が予め設定された閾値未満となった場合、又は、一定回数以上繰り返した場合、処理を終了する。そして、最終的に得られたパラメータα,β,σを、通行モデルに使用する。このように、通行モデルのパラメータを求めることで、通行人数をより正確に推定することが可能となる。   Specifically, the parameters α, β, and σ are determined by using the EM algorithm. First, the occupancy number estimating unit 14 estimates the number of passing persons using the passing model as described above using the provisional values of the parameters α, β, and σ determined in advance (E step). Next, the occupant number estimating unit 14 uses the number of passing persons estimated in the E step and obtains parameters α, β, and σ by regression analysis (least square method) (M step). Then, using the parameters α, β, σ obtained in the M steps as new provisional values, the number of people passing is estimated by the passing model (E step), and the parameters α, β,. σ is obtained (M step). As described above, the E step and the M step are repeatedly executed, and when the amount of change in the parameters α, β, and σ is less than a preset threshold value or when the change is repeated a predetermined number of times or more, the process is terminated. Then, finally obtained parameters α, β, and σ are used in the traffic model. Thus, by obtaining the parameters of the traffic model, it is possible to more accurately estimate the number of traffic people.

(その他の変形例)
上述した実施形態及び変形例では、退室人数を推定する場合について説明したが、扉22が、人が検知された場合に開く自動ドアである場合、入室人数も推定するようにしてもよい。例えば、在室人数推定部14は、部屋20の入室側(外側)の人が検知されて扉22が開いた場合、そのときの扉開時間間隔に基づいて入室人数を推定し、部屋20の退室側(内側)の人が検知されて扉22が開いた場合、そのときの扉開時間間隔に基づいて退室人数を推定してもよい。そして、在室人数推定部14は、推定された入室人数と退室人数とから部屋20の在室人数を求めてもよい。このように、入退室管理装置10は、退室人数を推定する退室人数推定装置として機能してもよいし、入室人数を推定する入室人数推定装置として機能してもよい。
(Other variations)
In the embodiment and the modification described above, the case of estimating the number of people leaving the room has been described. However, when the door 22 is an automatic door that opens when a person is detected, the number of people entering the room may also be estimated. For example, when a person on the entrance side (outside) of the room 20 is detected and the door 22 is opened, the occupancy number estimation unit 14 estimates the number of people entering the room 20 based on the door opening time interval at that time. When a person on the exit side (inside) is detected and the door 22 is opened, the number of persons leaving the room may be estimated based on the door opening time interval. Then, the occupancy number estimation unit 14 may obtain the occupancy number of the room 20 from the estimated number of people entering and leaving the room. Thus, the entrance / exit management device 10 may function as an exit number estimation device that estimates the number of exits, or may function as an entrance number estimation device that estimates the number of entrances.

上述した入退室管理装置10は図示しないCPU等のプロセッサを備えている。プロセッサは、図示しないメモリに記憶されたプログラムを実行することにより、入退室管理部12及び在室人数推定部14のそれぞれの機能を実現する。上記プログラムは、CDやDVD等の記録媒体を介して又はネットワーク等の通信経路を介してハードディスクドライブ(HDD)等の記憶装置に記憶される。なお、上記プログラムはハードディスクドライブ等の記憶装置に予め記憶されていてもよい。ハードディスクドライブ等の記憶装置に記憶されたプログラムがRAM等のメモリに読み出されてCPU等のプロセッサによって実行されることにより、入退室管理部12及び在室人数推定部14のそれぞれの機能が実現される。   The entrance / exit management device 10 described above includes a processor such as a CPU (not shown). A processor implement | achieves each function of the entrance / exit management part 12 and the occupancy number estimation part 14 by running the program memorize | stored in the memory which is not shown in figure. The program is stored in a storage device such as a hard disk drive (HDD) via a recording medium such as a CD or DVD, or via a communication path such as a network. The program may be stored in advance in a storage device such as a hard disk drive. Each function of the entrance / exit management unit 12 and the occupancy number estimation unit 14 is realized by a program stored in a storage device such as a hard disk drive being read into a memory such as a RAM and executed by a processor such as a CPU. Is done.

10 入退室管理装置、12 入退室管理部、14 在室人数推定部、16 出力部、20 部屋、22 扉、24 電気錠、26 扉開閉センサ、28 認証装置。
DESCRIPTION OF SYMBOLS 10 Entrance / exit management apparatus, 12 Entrance / exit management part, 14 Current occupancy estimation part, 16 Output part, 20 rooms, 22 Doors, 24 Electric lock, 26 Door opening / closing sensor, 28 Authentication apparatus.

Claims (8)

部屋に入室する人を検知する検知手段と、
前記部屋からの退室時に前記部屋の扉が開いている時間間隔に基づき、前記扉の各開閉時点における退室人数を推定し、前記検知手段によって検知された入室人数と前記推定された退室人数とに基づき、前記扉の各開閉時点における前記部屋の在室人数を求める在室人数推定手段と、
を有し、
前記在室人数推定手段は、所定期間内の退室人数の合計が、前記所定期間内の入室人数の合計と等しくなるように、退室人数の合計を前記扉の各開閉時点における退室人数に割り振り、退室人数の割り振りが異なる複数の組み合わせのそれぞれについて、前記扉の各開閉時点における入室人数と退室人数とから前記扉の各開閉時点における通行人数を求め、前記扉の各開閉時点における通行人数と、前記扉の各開閉時点において前記扉が開いている時間間隔とに基づいて尤度を求め、尤度が最大となる組み合わせの入室人数と退室人数とから求められる在室人数を推定値として出力する、
ことを特徴とする入退室管理装置。
Detection means for detecting a person entering the room;
Based on the time interval when the door of the room is open when leaving the room, the number of people leaving the room at each opening and closing time is estimated, and the number of people entering the room detected by the detection means and the estimated number of people leaving the room Based on the occupancy number estimation means for determining the occupancy number of the room at each opening and closing time of the door,
I have a,
The occupancy number estimating means allocates the total number of people leaving the room at each opening and closing time of the door so that the total number of people leaving the room within a predetermined period is equal to the total number of people entering the room within the predetermined period. For each of a plurality of combinations with different allocation of the number of people leaving the room, the number of people passing at the time of opening and closing the door is determined from the number of people entering and leaving the door at the time of opening and closing the door, and the number of people passing at the time of opening and closing the door, The likelihood is calculated based on the time interval at which the door is opened at each opening and closing time of the door, and the number of people in the room obtained from the number of entering and leaving the combination with the maximum likelihood is output as an estimated value. ,
An entrance / exit management device characterized by that.
請求項に記載の入退室管理装置であって、
前記検知手段は、前記部屋に入室する人を認証して入室人数を検知し、
前記在室人数推定手段は、前記複数の組み合わせのそれぞれについて、入室人数と割り振られた退室人数とから前記扉の各開閉時点における在室人数を求め、前記検知手段の認証結果に基づき、同一人物の重複の無い入室人数の合計を前記複数の組み合わせのそれぞれについて求め、前記複数の組み合わせのうち、前記扉の各開閉時点における在室人数が、同一人物の重複の無い入室人数の合計以下となる組み合わせを対象にして、尤度が最大となる組み合わせを決定し、前記決定された組み合わせの在室人数を推定値として出力する、
ことを特徴とする入退室管理装置。
The entrance / exit management device according to claim 1 ,
The detection means detects the number of people entering the room by authenticating the person entering the room,
The occupancy number estimating means obtains the occupancy number at each opening / closing time of the door from the number of entering rooms and the number of leaving rooms assigned for each of the plurality of combinations, and based on the authentication result of the detecting means, the same person The total number of occupants without duplication is calculated for each of the plurality of combinations, and the number of occupants at each opening and closing time of the door is equal to or less than the total number of occupants without duplication of the same person. For the combination, determine the combination with the maximum likelihood, and output the number of people in the determined combination as an estimated value.
An entrance / exit management device characterized by that.
部屋に入室する人を検知する検知手段と、
前記部屋からの退室時に前記部屋の扉が開いている時間間隔に基づき、前記扉の各開閉時点における退室人数を推定し、前記検知手段によって検知された入室人数と前記推定された退室人数とに基づき、前記扉の各開閉時点における前記部屋の在室人数を求める在室人数推定手段と、
を有し、
前記在室人数推定手段は、前記扉の各開閉時点において扉が開いている時間間隔と通行人数とに基づいて求められる尤度が最大となるように、前記扉の各開閉時点における退室人数を決定し、入室人数と決定された退室人数とから求められる在室人数を推定値として出力する、
ことを特徴とする入退室管理装置。
Detection means for detecting a person entering the room;
Based on the time interval when the door of the room is open when leaving the room, the number of people leaving the room at each opening and closing time is estimated, and the number of people entering the room detected by the detection means and the estimated number of people leaving the room Based on the occupancy number estimation means for determining the occupancy number of the room at each opening and closing time of the door,
Have
The occupancy estimation means calculates the number of people leaving the room at each opening and closing time of the door so that the likelihood obtained based on the time interval at which the door is open and the number of passing persons at the time of opening and closing the door is maximized. The number of people in the room determined from the number of people entering and the number of people left is output as an estimated value.
An entrance / exit management device characterized by that.
請求項から請求項のいずれか一項に記載の入退室管理装置であって、
前記在室人数推定手段は、人が前記扉の通行に要する時間の長さに通行人数を乗算して得られる値と、前記扉の開閉に要する時間との合計を、前記扉が開いている時間間隔と定義する通行モデルを用いて、尤度を求める、
ことを特徴とする入退室管理装置。
A entry and exit management apparatus according to any one of claims 1 to 3,
The occupant number estimating means is configured such that the door is the sum of a value obtained by multiplying the length of time required for a person to pass through the number of passing persons and the time required for opening and closing the door. Using a traffic model defined as a time interval, find the likelihood.
An entrance / exit management device characterized by that.
請求項に記載の入退室管理装置であって、
前記在室人数推定手段は、通行人数と前記扉が開いている時間間隔とに基づいて、人が前記扉の通行に要する時間と前記扉の開閉に要する時間とを新たに求め、前記新たに求められた値と前記通行モデルとを用いて、尤度を求める、
ことを特徴とする入退室管理装置。
The entrance / exit management device according to claim 4 ,
The occupant number estimating means newly determines the time required for the person to pass through the door and the time required for opening and closing the door based on the number of people passing and the time interval at which the door is open. Using the obtained value and the traffic model, the likelihood is obtained.
An entrance / exit management device characterized by that.
請求項から請求項のいずれか一項に記載の入退室管理装置であって、
前記在室人数推定手段は、入室人数及び退室人数を増加させ、増加後の入室人数と退室人数とに基づき通行人数を求め、増加後の通行人数と前記扉が開いている時間間隔とに基づいて尤度を求め、増加後の尤度が増加前の尤度よりも高い場合、増加後の入室人数と退室人数とから求められる在室人数を推定値として出力する、
ことを特徴とする入退室管理装置。
A entry and exit management apparatus according to any one of claims 1 to 4,
The occupancy number estimating means increases the number of people entering and leaving the room, obtains the number of people passing by based on the number of people entering the room and the number of people leaving the room, and based on the increased number of people passing and the time interval at which the door is open. If the likelihood after the increase is higher than the likelihood before the increase, the number of people in the room calculated from the number of people entering and leaving after the increase is output as an estimated value.
An entrance / exit management device characterized by that.
コンピュータに、
部屋に入室する人を検知するステップと、
前記部屋からの退室時に前記部屋の扉が開いている時間間隔に基づき、前記扉の各開閉時点における退室人数を推定し、検知された入室人数と前記推定された退室人数とに基づき、前記扉の各開閉時点における前記部屋の在室人数を求めるステップと、
を実行させ
前記在室人数を求めるステップでは、所定期間内の退室人数の合計が、前記所定期間内の入室人数の合計と等しくなるように、退室人数の合計を前記扉の各開閉時点における退室人数に割り振り、退室人数の割り振りが異なる複数の組み合わせのそれぞれについて、前記扉の各開閉時点における入室人数と退室人数とから前記扉の各開閉時点における通行人数を求め、前記扉の各開閉時点における通行人数と、前記扉の各開閉時点において前記扉が開いている時間間隔とに基づいて尤度を求め、尤度が最大となる組み合わせの入室人数と退室人数とから求められる在室人数を推定値として出力する、
ことを特徴とするプログラム。
On the computer,
Detecting a person entering the room;
Based on the time interval when the door of the room is open when leaving the room, the number of persons leaving the door at each opening and closing time is estimated, and based on the detected number of persons entering the room and the estimated number of persons leaving the door, the door Determining the number of people in the room at each opening and closing time of
Was executed,
In the step of obtaining the number of people in the room, the total number of people leaving the room is allocated to the number of people leaving the room at each opening and closing time of the door so that the total number of people leaving the room within a predetermined period is equal to the total number of people who have entered the room during the predetermined period. For each of a plurality of combinations with different allocations of the number of people leaving the room, the number of people passing at the time of opening and closing the door is determined from the number of people entering and leaving the door at the time of opening and closing the door. The likelihood is obtained based on the time interval at which the door is opened at each opening and closing time of the door, and the number of people in the room obtained from the number of entering and leaving the combination with the maximum likelihood is output as an estimated value. To
A program characterized by that.
コンピュータに、  On the computer,
部屋に入室する人を検知するステップと、  Detecting a person entering the room;
前記部屋からの退室時に前記部屋の扉が開いている時間間隔に基づき、前記扉の各開閉時点における退室人数を推定し、検知された入室人数と前記推定された退室人数とに基づき、前記扉の各開閉時点における前記部屋の在室人数を求めるステップと、  Based on the time interval when the door of the room is open when leaving the room, the number of persons leaving the door at each opening and closing time is estimated, and based on the detected number of persons entering the room and the estimated number of persons leaving the door, the door Determining the number of people in the room at each opening and closing time of
を実行させ、  And execute
前記在室人数を求めるステップでは、前記扉の各開閉時点において扉が開いている時間間隔と通行人数とに基づいて求められる尤度が最大となるように、前記扉の各開閉時点における退室人数を決定し、入室人数と決定された退室人数とから求められる在室人数を推定値として出力する、  In the step of obtaining the number of people in the room, the number of people leaving the room at each opening and closing time of the door is maximized so that the likelihood obtained based on the time interval at which the door is open and the number of passing persons at the time of opening and closing the door is maximized. , And output the estimated number of people in the room determined from the number of people entering and the number of people leaving the room,
ことを特徴とするプログラム。  A program characterized by that.
JP2013027128A 2013-02-14 2013-02-14 Entrance / exit management device and program Expired - Fee Related JP5875541B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2013027128A JP5875541B2 (en) 2013-02-14 2013-02-14 Entrance / exit management device and program

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2013027128A JP5875541B2 (en) 2013-02-14 2013-02-14 Entrance / exit management device and program

Publications (2)

Publication Number Publication Date
JP2014157422A JP2014157422A (en) 2014-08-28
JP5875541B2 true JP5875541B2 (en) 2016-03-02

Family

ID=51578273

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2013027128A Expired - Fee Related JP5875541B2 (en) 2013-02-14 2013-02-14 Entrance / exit management device and program

Country Status (1)

Country Link
JP (1) JP5875541B2 (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP6489285B2 (en) * 2016-04-04 2019-03-27 三菱電機株式会社 Entrance / exit management system
JPWO2020152760A1 (en) * 2019-01-21 2021-12-16 株式会社tsumug.lab Permit control system
KR102491161B1 (en) * 2020-11-17 2023-01-20 조선대학교 산학협력단 Presence estimation device

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0535952A (en) * 1991-01-31 1993-02-12 Alpha:Kk Room entering/leaving controlling device
JP2011144505A (en) * 2010-01-12 2011-07-28 Panasonic Electric Works Co Ltd Management system

Also Published As

Publication number Publication date
JP2014157422A (en) 2014-08-28

Similar Documents

Publication Publication Date Title
JP4924713B2 (en) Entrance / exit management program, entrance / exit management method, and entrance / exit management device
US10043325B2 (en) Access control for areas with multiple doors
US10026246B2 (en) Adaptive access control for areas with multiple doors
JP5592726B2 (en) Entrance / exit management system and method
JP5040381B2 (en) Entrance / exit management device
US20230401914A1 (en) Information processing apparatus, information processing system, and information processing method
JP5875541B2 (en) Entrance / exit management device and program
US20220415105A1 (en) Information processing apparatus, information processing system, and information processing method
JP2014005993A (en) Air conditioning control support device, air conditioning control support method, and program therefor
WO2012036692A1 (en) Security device with security image update capability
US11164414B2 (en) System and method for providing secure access
JP2019159974A (en) Authentication device, authentication method and authentication program
KR101395675B1 (en) Access control system and method
US9256996B2 (en) Method and system for training users related to a physical access control system
JP2013114553A (en) Room entry/exit management system and room entry/exit management program
JP6250494B2 (en) Portable storage device, action history management system, and program
JP5029292B2 (en) Entrance / exit management system
JP6988529B2 (en) Piggybacking detector
TWI332628B (en)
JP2013194467A (en) Automatic door control device and control method
JP2011047149A (en) Access management system, and environmental control system
JP5929225B2 (en) Entrance / exit management system
JP2006070653A (en) Security passing controller
JP5780113B2 (en) Entrance / exit management system
JP2011179250A (en) System for managing entrance into/exit from room

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20141210

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20150925

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20151020

A521 Request for written amendment filed

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20151202

TRDD Decision of grant or rejection written
A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

Effective date: 20151222

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20160119

R150 Certificate of patent or registration of utility model

Ref document number: 5875541

Country of ref document: JP

Free format text: JAPANESE INTERMEDIATE CODE: R150

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250

S533 Written request for registration of change of name

Free format text: JAPANESE INTERMEDIATE CODE: R313533

R350 Written notification of registration of transfer

Free format text: JAPANESE INTERMEDIATE CODE: R350

LAPS Cancellation because of no payment of annual fees