JP5241461B2 - Passenger detection device for passenger conveyor - Google Patents

Passenger detection device for passenger conveyor Download PDF

Info

Publication number
JP5241461B2
JP5241461B2 JP2008312323A JP2008312323A JP5241461B2 JP 5241461 B2 JP5241461 B2 JP 5241461B2 JP 2008312323 A JP2008312323 A JP 2008312323A JP 2008312323 A JP2008312323 A JP 2008312323A JP 5241461 B2 JP5241461 B2 JP 5241461B2
Authority
JP
Japan
Prior art keywords
distance
passenger
data string
congestion degree
weighting
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.)
Active
Application number
JP2008312323A
Other languages
Japanese (ja)
Other versions
JP2010132444A (en
Inventor
健太郎 林
敬秀 平井
広幸 蔦田
賢 新土井
浩一 竹内
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
Original Assignee
Mitsubishi Electric Corp
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 filed Critical Mitsubishi Electric Corp
Priority to JP2008312323A priority Critical patent/JP5241461B2/en
Publication of JP2010132444A publication Critical patent/JP2010132444A/en
Application granted granted Critical
Publication of JP5241461B2 publication Critical patent/JP5241461B2/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Escalators And Moving Walkways (AREA)

Description

この発明は、乗客コンベアにおける乗降口付近の人物や物体の状態を検知する乗客コンベアの乗客検出装置に関する。   The present invention relates to a passenger detection device for a passenger conveyor that detects the state of a person or an object in the vicinity of a passenger entrance on the passenger conveyor.

従来、乗客コンベアにおける乗降客の検出方法として、例えば特許文献１に記載されるように、車椅子利用者に赤外線を照射し、利用者に反射した光を赤外線カメラによって検出し、この反射光のパターンを学習、更新する。そして、学習したパターンと入力とを比較することにより車椅子利用者を検出するようにしたものがあった。   Conventionally, as a method for detecting passengers on a passenger conveyor, as described in Patent Document 1, for example, a wheelchair user is irradiated with infrared light, and light reflected on the user is detected by an infrared camera, and this reflected light pattern is detected. Learn and update And there existed what detected the wheelchair user by comparing the learned pattern and input.

特開平０５−７００７５号公報JP 05-70075 A

従来の装置では、複数の赤外線ビームを赤外線カメラで計測することにより、数十点以上の比較的密な計測点を得て、その情報から乗り場の状態（車椅子利用者）を検知していた。このような密な計測点を得ることできれば、計測精度を向上させることができる。
しかしながら、密な計測点を得るためには、スキャン型レーザ測距センサであればモータ等の可動部が必要であり、また、カメラ方式であってもカメラを別の位置に設置しなければならないなど、装置、設置コストが高くなり、また構成が複雑になることによって耐障害性が低くなるなどの課題があった。 In the conventional apparatus, by measuring a plurality of infrared beams with an infrared camera, a relatively dense measurement point of several tens or more is obtained, and the state of the landing (wheelchair user) is detected from the information. If such dense measurement points can be obtained, the measurement accuracy can be improved.
However, in order to obtain dense measurement points, a movable part such as a motor is necessary for a scanning laser distance measuring sensor, and the camera must be installed at a different position even in the camera system. As a result, there are problems such as an increase in device and installation costs and a reduction in fault tolerance due to a complicated configuration.

この発明は上記のような課題を解決するためになされたもので、少ない計測点であっても効率よくかつ正確に検知することができる乗客コンベアの乗客検出装置を得ることを目的とする。   The present invention has been made to solve the above-described problems, and an object thereof is to obtain a passenger detection device for a passenger conveyor that can efficiently and accurately detect even a small number of measurement points.

この発明に係る乗客コンベアの乗客検出装置は、測距センサより複数の方向へビームを放射して乗客コンベアにおける乗降口付近の乗客への距離を計測し、計測結果を距離データ列として出力する距離データ列取得手段と、距離データ列に基づいて、乗降口付近の乗客のデータ密度を求めると共に、乗降口付近における領域毎の重み付け値を示す事前知識を用いて、求めたデータ密度に対して重み付けを行って乗客の混雑度を算出する混雑度算出手段とを備えたものである。   The passenger detection device for a passenger conveyor according to the present invention measures the distance to passengers near the entrance / exit on the passenger conveyor by radiating beams from a distance measuring sensor in a plurality of directions, and outputs the measurement results as a distance data string Based on the data string acquisition means and the distance data string, the data density of passengers near the entrance / exit is obtained, and the prior data indicating the weight value for each area in the vicinity of the entrance / exit is used to weight the obtained data density And a congestion degree calculating means for calculating the degree of congestion of passengers.

この発明の乗客コンベアの乗客検出装置は、乗降口付近における領域毎の重み付け値を示す事前知識を用いて、距離データ列から求めたデータ密度に対して重み付けを行って乗客の混雑度を算出するようにしたので、少ない計測点であっても効率よくかつ正確に乗客の混雑度を検知することができる。   The passenger detection device for a passenger conveyor according to the present invention calculates the degree of congestion of passengers by weighting the data density obtained from the distance data string, using prior knowledge indicating the weighting value for each region in the vicinity of the entrance / exit. Since it did in this way, even if it is few measurement points, a passenger's congestion degree can be detected efficiently and correctly.

実施の形態１．
図１は、この発明の実施の形態１による乗客コンベアの乗客検出装置を示す構成図である。
図において、乗客コンベアの乗客検出装置は、距離データ列取得手段１、事前知識２、混雑度算出手段３を備えている。距離データ列取得手段１は、エスカレータ４のインレット部に設けられた測距センサ５を含んでいる。この測距センサ５は、例えば、各ビーム間の角度が５度程度と、粗い密度で水平方向にビームを放射して測距するセンサであり、乗降客（図示せず）の体の一部に反射するビームを捉えて、反射位置までの距離を計測する。尚、測距センサ５の設置個数は一つだけでなく、例えば乗降口の両側のインレット部に設ける等、複数個であってもよい。距離データ列取得手段１は、測距センサ５で得られた複数地点（例えば数点〜十数点程度）の距離データを距離データ列として取得し、この距離データ列を出力する機能部である。尚、距離データ列取得手段１における「距離データ列取得１０１」や、混雑度算出手段３における「データ密度３０１」といったブロックは、各手段における処理やデータを示しており、これらについては後述する。 Embodiment 1 FIG.
1 is a block diagram showing a passenger detection device for a passenger conveyor according to Embodiment 1 of the present invention.
In the figure, a passenger detection device for a passenger conveyor includes distance data string acquisition means 1, prior knowledge 2, and congestion degree calculation means 3. The distance data string acquisition means 1 includes a distance measuring sensor 5 provided in the inlet portion of the escalator 4. The distance measuring sensor 5 is a sensor that radiates a beam in a horizontal direction with a rough density, for example, with an angle between the beams of about 5 degrees, and is a part of the body of a passenger (not shown). Measure the distance to the reflection position by capturing the reflected beam. The number of distance measuring sensors 5 is not limited to one, and may be plural, for example, provided at the inlet portions on both sides of the entrance / exit. The distance data string acquisition unit 1 is a functional unit that acquires distance data of a plurality of points (for example, about several to tens of points) obtained by the distance measuring sensor 5 as a distance data string and outputs the distance data string. . Blocks such as “distance data string acquisition 101” in the distance data string acquisition unit 1 and “data density 301” in the congestion degree calculation unit 3 indicate processing and data in each unit, which will be described later.

事前知識２は、予め求めた測距センサ５の各領域毎の重み付け値や、時系列変化３０３における時間間隔や測距センサ５の距離に対応した重み付け値といったデータである。混雑度算出手段３は、距離データ列取得手段１から出力された距離データ列からデータ密度３０１や時系列変化３０３を求め、更に、事前知識２に基づいて重みづけ総和３０２を演算して、エスカレータ４の乗降口付近の乗降客の混雑度を算出する機能部である。   The prior knowledge 2 is data such as a weight value for each area of the distance measurement sensor 5 obtained in advance, a time value in the time series change 303, or a weight value corresponding to the distance of the distance measurement sensor 5. The congestion degree calculation means 3 obtains the data density 301 and the time series change 303 from the distance data string output from the distance data string acquisition means 1, further calculates the weighted sum 302 based on the prior knowledge 2, and calculates the escalator. 4 is a functional unit for calculating the degree of congestion of passengers near the entrance / exit of No. 4;

尚、乗客コンベアの乗客検出装置はコンピュータを用いて実現され、距離データ列取得手段１における演算部と混雑度算出手段３とは、それぞれの機能に対応するソフトウェアと、これを実行するためのＣＰＵやメモリ等のハードウェアから構成されている。また、事前知識２は、ハードディスク装置といった不揮発性メモリに格納されている。尚、距離データ列取得手段１や混雑度算出手段３は専用のハードウェアで構成してもよい。更に、上記例では、乗客コンベアとしてエスカレータ４を挙げているが、これ以外にも例えば動く歩道等でも同様に適用可能である。   The passenger detection device of the passenger conveyor is realized by using a computer. The calculation unit and the congestion degree calculation unit 3 in the distance data string acquisition unit 1 are software corresponding to each function and a CPU for executing the software. And hardware such as memory. The prior knowledge 2 is stored in a nonvolatile memory such as a hard disk device. The distance data string acquisition unit 1 and the congestion degree calculation unit 3 may be configured with dedicated hardware. Furthermore, although the escalator 4 is mentioned as a passenger conveyor in the above example, other than this, for example, a moving sidewalk or the like can be similarly applied.

次に、実施の形態１の乗客コンベアの乗客検出装置の動作について説明する。
先ず、距離データ列取得手段１において、測距センサ５が複数地点の距離データを取得し、これを距離データ列として出力する（図１における１０１）。
混雑度算出手段３における混雑度計算３０４は、距離データ列取得手段１から出力された距離データ列に基づくデータ密度３０１、重みづけ総和３０２、時系列変化３０３の値を基に計算される。
データ密度３０１は、図２に示す測距センサ５のビーム範囲である扇状領域内のデータにある適切な重みを掛け、総和をとったものである。この場合の適切な重みとは、以下の考え方により事前知識２として決定することができる。例えば、図２に示すように複数の測距センサ５が同一エリアを計測するような場合、エリアによってビームの密度が異なるため、これを補正することができる。図２には二つの測距センサ５があり、それぞれ扇型のエリアを計測しているとする。領域Ａは二つのセンサのエリアが重なる部分であり、領域Ｂは一方のセンサのみ計測するエリアである。領域Ａと領域Ｂで同じ重みで計測点数の総和を取ると、同じ大きさの物体であったとしても計測点数が異なることになる。そこで、領域Ａには領域Ｂの半分の重みを設定する。このようにすることで、同じ物体であれば同じデータ密度が得られることが期待できる。 Next, operation | movement of the passenger detection apparatus of the passenger conveyor of Embodiment 1 is demonstrated.
First, in the distance data string acquisition means 1, the distance measuring sensor 5 acquires distance data of a plurality of points and outputs it as a distance data string (101 in FIG. 1).
The congestion degree calculation 304 in the congestion degree calculation means 3 is calculated based on the values of the data density 301, weighted sum 302, and time series change 303 based on the distance data string output from the distance data string acquisition means 1.
The data density 301 is obtained by multiplying the data in the fan-shaped area, which is the beam range of the distance measuring sensor 5 shown in FIG. An appropriate weight in this case can be determined as prior knowledge 2 based on the following concept. For example, as shown in FIG. 2, when a plurality of distance measuring sensors 5 measure the same area, the beam density differs depending on the area, and this can be corrected. In FIG. 2, it is assumed that there are two distance measuring sensors 5, each measuring a fan-shaped area. Region A is a portion where the areas of two sensors overlap, and region B is an area where only one sensor is measured. If the total number of measurement points is taken with the same weight in the region A and the region B, the number of measurement points will be different even if the objects have the same size. Therefore, half the weight of area B is set for area A. By doing so, it can be expected that the same data density can be obtained for the same object.

このように補正した後に、エスカレータ特有の条件を加味し、例えば図３のように、測距センサ５から近い扇状領域内に存在するデータ点には大きな重みを掛け、それ以外の領域にあり、それより大きな扇状領域内に存在するデータ点には小さな重みを掛ける。これは、エスカレータ４の乗降口に近い領域に人物が存在する場合は危険度が高いという事前知識と、乗降口が混雑してきた場合には多数の人物や物体がセンサの直前に存在し、従って重みの大きな領域に点が集中する傾向があるという知識による。このようにすることで、危険度が高まるのに相関して変動する数値を作成できる。   After correcting in this way, taking into consideration the conditions specific to the escalator, for example, as shown in FIG. 3, the data points existing in the fan-shaped area close to the distance measuring sensor 5 are heavily weighted, and are in other areas, A small weight is applied to a data point existing in a larger fan-shaped area. This is because prior knowledge that the degree of danger is high when a person is present in an area close to the entrance / exit of the escalator 4, and when the entrance / exit is congested, a large number of persons and objects exist immediately in front of the sensor. It is based on the knowledge that points tend to concentrate in areas with large weights. By doing in this way, the numerical value which fluctuates in correlation with the increase in the risk can be created.

次に、時系列変化３０３は、図４に示す方法によって計算する。グラフの中で、横軸はビーム、縦軸は距離を表す。実線は現在時刻の距離分布、破線は直前の時刻の距離分布を示す。このとき、同じビーム間で、直前と現在時刻の距離値の差分の絶対値を計算し、また直前あるいは現在時刻の距離がどの重み領域にあるかによってその絶対値に重みを掛ける。例えば、図４の（ア）においては、距離が重み大の領域にあるので、（ア）の差分量に大きな重みを掛ける。逆に（ウ）の領域は、距離が重み小の領域にあるので、（ウ）の差分量に小さな重みを掛ける。（イ）の領域は、直前の距離値が重み中の範囲にあり、現在の距離値が重み小の範囲にあり、いずれを選択すればよいかであるが、例えば重みの大きな方を優先するといった考えを適用すれば、大きな重みを掛ける。これら重み係数は事前知識であるため、予め決められた値を事前知識２として保持しておく。尚、縦軸の標準偏差については実施の形態２で説明する。   Next, the time series change 303 is calculated by the method shown in FIG. In the graph, the horizontal axis represents the beam, and the vertical axis represents the distance. The solid line indicates the distance distribution at the current time, and the broken line indicates the distance distribution at the immediately preceding time. At this time, the absolute value of the difference between the distance value between the previous time and the current time is calculated between the same beams, and the absolute value is weighted depending on which weight region the distance between the previous time or the current time is. For example, in (A) of FIG. 4, since the distance is in a region having a large weight, a large weight is applied to the difference amount of (A). On the contrary, since the area (c) is in an area where the distance is small, a small weight is applied to the difference amount of (c). In the area (b), the immediately preceding distance value is in the weighting range and the current distance value is in the small weight range, which should be selected. For example, priority is given to the larger weight value. If such an idea is applied, large weight is applied. Since these weighting coefficients are prior knowledge, a predetermined value is held as prior knowledge 2. The standard deviation on the vertical axis will be described in the second embodiment.

混雑度算出手段３は、上記全てのビームについて同様の計算を行い、重みづけ総和３０２を計算する。重みの値と領域の区分は、事前知識２による。距離が近い部分の重みが大きいのは、近い距離の精度は高く、遠い場合には精度が低いようなセンサ特性である場合を想定している。精度の高い測距センサ５を用いればそのような必要はないが、その場合はコストが高くなる可能性がある。   The congestion degree calculation means 3 performs the same calculation for all the beams, and calculates the weighted sum 302. The weight value and the area classification are based on prior knowledge 2. The reason why the weight of the portion close to the distance is large is based on the sensor characteristic in which the accuracy of the close distance is high and the accuracy is low when the distance is long. If the distance measuring sensor 5 with high accuracy is used, such a need is not required, but in that case, the cost may increase.

このようにして得られたデータ密度３０１および時系列変化３０３を用いて、両者の重みづけ総和３０２を計算し、この重みづけ総和３０２に基づいて混雑度計算３０４を行い、結果を出力３０５する。両者の重みづけは設置場所の利用状況や検知したい状況を勘案すればよい。例えば、混雑することは少ないが、インレット付近に立ち止まる人物が多く、こうした事象を主に検知したいのであればデータ密度３０１に対する重みを大きくする。逆に立ち止まる人物は少ないが、混雑する状況が多い場合（駅など）には、時系列変化３０３の重みを大きくする。このようにすることで、設置状況に応じて検知した事象を効果的に検知することが可能となる。   Using the data density 301 and the time series change 303 obtained in this way, a weighted sum 302 is calculated, a congestion degree calculation 304 is performed based on the weighted sum 302, and the result is output 305. The weighting of both may take into account the usage situation of the installation location and the situation to be detected. For example, although there is little crowding, there are many people who stop near the inlet, and if it is desired to mainly detect such an event, the weight for the data density 301 is increased. On the contrary, when there are few people who stop, but there are many crowded situations (such as a station), the weight of the time series change 303 is increased. By doing in this way, it becomes possible to detect effectively the phenomenon detected according to the installation situation.

以上のように、実施の形態１の乗客コンベアの乗客検出装置によれば、測距センサより複数の方向へビームを放射して乗客コンベアにおける乗降口付近の乗客への距離を計測し、この計測結果を距離データ列として出力する距離データ列取得手段と、距離データ列に基づいて、乗降口付近の乗客のデータ密度を求めると共に、乗降口付近における領域毎の重み付け値を示す事前知識を用いて、求めたデータ密度に対して重み付けを行って乗客の混雑度を算出する混雑度算出手段とを備えたので、少ない計測点であっても効率よくかつ正確に乗客の混雑度を検知することができる。   As described above, according to the passenger detection device for a passenger conveyor of the first embodiment, the distance sensor emits a beam in a plurality of directions to measure the distance to passengers near the entrance / exit on the passenger conveyor. Based on distance data string acquisition means for outputting the result as a distance data string, and determining the data density of passengers near the entrance / exit based on the distance data string, and using prior knowledge indicating weight values for each area near the entrance / exit In addition, since it is provided with a congestion degree calculation means for calculating the passenger congestion degree by weighting the obtained data density, it is possible to detect the passenger congestion degree efficiently and accurately even with a small number of measurement points. it can.

また、実施の形態１の乗客コンベアの乗客検出装置によれば、事前知識として測距センサの測距距離に対応した重み付け値を時系列変化に対する重み付け値として有し、混雑度算出手段は、距離データの時系列変化に対して、時系列変化がどの距離で発生したかに基づいて重み付け値で重み付けを行って乗客の混雑度を算出するようにしたので、設置状況に応じて検知した事象を効果的に検知することが可能となる。   In addition, according to the passenger detection device of the passenger conveyor of the first embodiment, as prior knowledge, the weighting value corresponding to the distance measured by the distance measuring sensor is used as the weighting value for the time series change, and the congestion degree calculation means Because the time series change of the data is weighted with a weighting value based on the distance at which the time series change occurred, the degree of congestion of the passenger is calculated, so the detected event according to the installation situation It becomes possible to detect effectively.

実施の形態２．
実施の形態２は、乗降客の転倒といった静止状態を検出するようにしたものである。
図５は、実施の形態２の乗客コンベアの乗客検出装置の構成図である。
図において、事前知識２ａは、実施の形態１の事前知識２の知識に加えて、測距センサ５の測距距離に応じた精度に基づく標準偏差値を有している。また、混雑度算出手段３ａは、実施の形態１における混雑度算出手段３の機能に加えて、事前知識２ａの標準偏差値を用いて転倒検知３０６を行うよう構成されている。他の各構成は実施の形態１と同様であるため、対応する部分に同一符号を付してその説明を省略する。 Embodiment 2. FIG.
In the second embodiment, a stationary state such as a passenger falling over is detected.
FIG. 5 is a configuration diagram of a passenger detection device for a passenger conveyor according to the second embodiment.
In the figure, the prior knowledge 2a has a standard deviation value based on the accuracy corresponding to the distance measured by the distance measuring sensor 5 in addition to the knowledge of the prior knowledge 2 of the first embodiment. Further, the congestion degree calculation means 3a is configured to perform the fall detection 306 using the standard deviation value of the prior knowledge 2a in addition to the function of the congestion degree calculation means 3 in the first embodiment. Since each other structure is the same as that of Embodiment 1, the same code | symbol is attached | subjected to a corresponding part and the description is abbreviate | omitted.

次に、実施の形態２の動作について説明する。
転倒検知３０６は、混雑度計算３０４で用いられた時系列変化３０３の考えを発展させる。実施の形態１と同様に図４に示した例を用いると、事前知識２ａにより測距センサ５の距離精度特性が近距離の場合に標準偏差が小さく、遠距離の場合に標準偏差が大きいということが分かっている場合、この標準偏差値により差分を正規化することによって静止状態（例えば、乗降客Ｃの転倒状態）であるかどうかを正確に計算することができる。
例えば、図４における（ア）領域の差分は、標準偏差が小さいため、小さい標準偏差で割り算して正規化する。逆に（ウ）領域の差分は、標準偏差が大きいため、大きい値で割り算して正規化する。各距離に応じた標準偏差値は事前に測距試験などで求めておき、事前知識２ａとして蓄えておく。標準偏差で割ることにより、いずれの距離値であっても静止状態であれば同じばらつきのノイズ値となる。 Next, the operation of the second embodiment will be described.
The fall detection 306 develops the idea of the time series change 303 used in the congestion degree calculation 304. If the example shown in FIG. 4 is used as in the first embodiment, the prior knowledge 2a indicates that the standard deviation is small when the distance accuracy characteristic of the distance measuring sensor 5 is short distance, and the standard deviation is large when the distance sensor is long distance. If it is known, it is possible to accurately calculate whether or not the vehicle is stationary (for example, the falling state of the passenger C) by normalizing the difference with the standard deviation value.
For example, the difference in the area (a) in FIG. 4 is normalized by dividing by a small standard deviation because the standard deviation is small. Conversely, since the difference in (c) region has a large standard deviation, it is normalized by dividing by a large value. A standard deviation value corresponding to each distance is obtained in advance by a distance measurement test or the like and stored as prior knowledge 2a. By dividing by the standard deviation, the noise value with the same variation can be obtained at any distance value in a stationary state.

混雑度算出手段３ａでは、全てのビームについて上記のように正規化された差分値を計算し、その二乗の総和を求める。この総和は、標準偏差のχ（カイ）２乗分布を成すと考えられる。各時刻で求めた総和の絶対値が、ある固定閾値以下にあれば、χ２乗分布から得られる確率で静止状態であると判定できる。そして、この固定閾値以下の状態が一定時間以上継続した場合に、転倒検知３０６の結果として乗降客Ｃが転倒状態であることを出力３０７する。この一定時間の値は、利用状況に応じて変更することも可能である。長くするほどノイズを拾いにくくなるが、同時生起確率は低下するため、見逃しが増える。この値は既に知られたχ２乗分布から求めることができるので、例えば顧客（本装置使用者）の要望通りの検知率を達成することが可能である。   In the congestion degree calculation means 3a, the difference values normalized as described above are calculated for all the beams, and the sum of the squares is obtained. This sum is considered to form a chi-square distribution of standard deviations. If the absolute value of the sum obtained at each time is equal to or less than a certain fixed threshold value, it can be determined that the object is in a stationary state with a probability obtained from the chi-square distribution. When the state equal to or lower than the fixed threshold value continues for a certain time or longer, the output 307 indicates that the passenger C is in a fallen state as a result of the fall detection 306. The value of the certain time can be changed according to the use situation. The longer it is, the more difficult it will be to pick up noise, but the probability of co-occurrence decreases, so overlooked increases. Since this value can be obtained from the already known chi-square distribution, for example, it is possible to achieve a detection rate as desired by the customer (user of this apparatus).

以上のような動作により、測距センサ５におけるビームの測距精度が悪くても精度よく静止状態を検知することができ、結果として少ないビームであっても静止状態、例えば、乗降客の転倒状態を検知することが可能となる。   With the operation as described above, it is possible to detect a stationary state with high accuracy even if the distance measurement accuracy of the beam in the distance measuring sensor 5 is poor. As a result, even with a small number of beams, the stationary state, for example, a passenger falling state Can be detected.

以上のように、実施の形態２の乗客コンベアの乗客検出装置によれば、事前知識として測距センサの測距距離に応じた精度に基づく標準偏差値を有し、混雑度算出手段は、標準偏差値を用いて各ビームの距離データの時系列変化を正規化すると共に、これら正規化したデータの総和を求めて所定の閾値と比較することにより乗客が静止状態であるかを判定するようにしたので、少ない計測点であっても効率よくかつ正確に乗客の転倒といった静止状態を検知することができる。   As described above, according to the passenger detection device for the passenger conveyor of the second embodiment, the standard deviation value based on the accuracy according to the distance measured by the distance measuring sensor is included as prior knowledge, and the congestion degree calculation means is a standard The deviation value is used to normalize the time series change of the distance data of each beam, and the total sum of these normalized data is obtained and compared with a predetermined threshold value to determine whether the passenger is stationary. Therefore, it is possible to detect a stationary state such as a passenger falling down efficiently and accurately even with a small number of measurement points.

この発明の実施の形態１による乗客コンベアの乗客検出装置を示す構成図である。It is a block diagram which shows the passenger detection apparatus of the passenger conveyor by Embodiment 1 of this invention. この発明の実施の形態１による乗客コンベアの乗客検出装置のデータ密度の領域による重み付けを示す説明図である。It is explanatory drawing which shows the weighting by the area | region of the data density of the passenger detection apparatus of the passenger conveyor by Embodiment 1 of this invention. この発明の実施の形態１による乗客コンベアの乗客検出装置のデータ密度の距離による重み付けを示す説明図である。It is explanatory drawing which shows the weighting by the distance of the data density of the passenger detection apparatus of the passenger conveyor by Embodiment 1 of this invention. この発明の実施の形態１による乗客コンベアの乗客検出装置の時系列変化の重み付けを示す説明図である。It is explanatory drawing which shows the weighting of the time series change of the passenger detection apparatus of the passenger conveyor by Embodiment 1 of this invention. この発明の実施の形態２による乗客コンベアの乗客検出装置を示す構成図である。It is a block diagram which shows the passenger detection apparatus of the passenger conveyor by Embodiment 2 of this invention.

符号の説明Explanation of symbols

１ 距離データ列取得手段、２，２ａ 事前知識、３，３ａ 混雑度算出手段、４ エスカレータ、５ 測距センサ、１０１ 距離データ列取得、３０１ データ密度、３０２ 重みづけ総和、３０３ 時系列変化、３０４ 混雑度計算、３０５，３０７ 出力、３０６ 転倒検知。   1 Distance data string acquisition means, 2, 2a Prior knowledge, 3, 3a Congestion degree calculation means, 4 Escalator, 5 Ranging sensor, 101 Distance data string acquisition, 301 Data density, 302 Weighted sum, 303 Time series change, 304 Congestion calculation, 305,307 output, 306 Fall detection.

Claims (2)

測距センサより複数の方向へビームを放射して乗客コンベアにおける乗降口付近の乗客への距離を計測し、当該計測結果を距離データ列として出力する距離データ列取得手段と、
前記距離データ列に基づいて、前記乗降口付近の乗客のデータ密度を求めると共に、当該乗降口付近における領域毎の重み付け値を示す事前知識を用いて、前記求めたデータ密度に対して重み付けを行って乗客の混雑度を算出する混雑度算出手段とを備えた乗客コンベアの乗客検出装置。 Distance data string acquisition means for radiating beams in a plurality of directions from the distance measuring sensor to measure the distance to passengers near the entrance / exit on the passenger conveyor, and outputting the measurement results as a distance data string;
Based on the distance data string, the data density of passengers near the entrance / exit is obtained, and weighting is performed on the obtained data density using prior knowledge indicating the weight value for each area in the vicinity of the entrance / exit. A passenger detection device for a passenger conveyor, comprising: a congestion degree calculating means for calculating a congestion degree of the passenger. 事前知識として測距センサの測距距離に対応した重み付け値を時系列変化に対する重み付け値として有し、混雑度算出手段は、距離データの時系列変化に対して、当該時系列変化がどの距離で発生したかに基づいて前記重み付け値で重み付けを行って乗客の混雑度を算出することを特徴とする請求項１記載の乗客コンベアの乗客検出装置。   As prior knowledge, it has a weighting value corresponding to the distance measured by the distance measuring sensor as a weighting value for the time series change, and the congestion degree calculating means determines at what distance the time series change is relative to the time series change of the distance data The passenger detection device for a passenger conveyor according to claim 1, wherein the passenger congestion level is calculated by weighting with the weighting value based on whether it occurs.
JP2008312323A 2008-12-08 2008-12-08 Passenger detection device for passenger conveyor Active JP5241461B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2008312323A JP5241461B2 (en) 2008-12-08 2008-12-08 Passenger detection device for passenger conveyor

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2008312323A JP5241461B2 (en) 2008-12-08 2008-12-08 Passenger detection device for passenger conveyor

Publications (2)

Publication Number Publication Date
JP2010132444A JP2010132444A (en) 2010-06-17
JP5241461B2 true JP5241461B2 (en) 2013-07-17

Family

ID=42344157

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2008312323A Active JP5241461B2 (en) 2008-12-08 2008-12-08 Passenger detection device for passenger conveyor

Country Status (1)

Country Link
JP (1) JP5241461B2 (en)

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102328870B (en) * 2010-07-12 2014-08-27 三菱电机株式会社 Falling detection device and passenger conveyor
KR101781225B1 (en) 2010-11-26 2017-09-22 미쓰비시 덴키 빌딩 테크노 서비스 가부시키 가이샤 Falling detector and passenger conveyor
CN102476774B (en) * 2010-11-26 2015-04-15 三菱电机株式会社 Tumble-detecting apparatus and passenger conveyor
JP5776289B2 (en) * 2010-11-26 2015-09-09 三菱電機株式会社 Fall detection device and passenger conveyor
KR20130028649A (en) * 2011-09-09 2013-03-19 미쓰비시 덴키 빌딩 테크노 서비스 가부시키 가이샤 Congestion level detection device and passenger conveyor
JP5377689B2 (en) * 2011-09-21 2013-12-25 斎藤 光正 LED lighting fixture with built-in standing wave radar
JP5689089B2 (en) * 2012-03-27 2015-03-25 三菱電機株式会社 Fall detection device and passenger conveyor
JP6321126B1 (en) * 2016-12-09 2018-05-09 東芝エレベータ株式会社 Passenger conveyor
WO2020105618A1 (en) * 2018-11-20 2020-05-28 日本電気株式会社 Passenger monitoring system and passenger monitoring method

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07157264A (en) * 1993-12-01 1995-06-20 Murata Mfg Co Ltd Conveyer
JP2000007267A (en) * 1998-06-17 2000-01-11 Hitachi Ltd Monitoring device for passenger conveyor
JP2000053361A (en) * 1998-08-12 2000-02-22 Hitachi Ltd Passenger monitoring device for man-conveyer
JP2003221179A (en) * 2002-01-31 2003-08-05 Ishikawajima Harima Heavy Ind Co Ltd Congestion detection system for variable speed moving footway
JP2008201509A (en) * 2007-02-19 2008-09-04 Mitsubishi Precision Co Ltd Person behavior discriminating system

Also Published As

Publication number Publication date
JP2010132444A (en) 2010-06-17

Similar Documents

Publication Publication Date Title
JP5241461B2 (en) Passenger detection device for passenger conveyor
US7760336B2 (en) Laser area sensor
JP6256531B2 (en) Object recognition processing device, object recognition processing method, and automatic driving system
US9664520B2 (en) Apparatus and method for generating assistance data with vertical access areas and predicted vertical movement models
US11536825B2 (en) Vehicle odometry and motion direction determination
US20140219505A1 (en) Pedestrian behavior predicting device and pedestrian behavior predicting method
JP4600929B2 (en) Stop low-speed vehicle detection device
JP5540217B2 (en) Laser scan sensor
JP5000928B2 (en) Object detection apparatus and method
JP2018022270A (en) Road shoulder detection method and road shoulder detection device
CN115331413A (en) Barrier gate rod early warning method and device, electronic equipment and storage medium
JP6704519B2 (en) Elevator equipment
JP5470762B2 (en) Object detection apparatus, object detection program, and object detection method
JP6396647B2 (en) Obstacle detection device and obstacle detection method
JP5993597B2 (en) Target moving object detection method
JP4893395B2 (en) Course prediction device and collision prediction device
JP6851111B2 (en) Railroad crossing obstacle detection device
JP7327315B2 (en) Abnormality detection method for infrastructure sensor device, and infrastructure sensor system
JP2012212407A (en) State determining device, state determining method and program
KR101755690B1 (en) Method for detecting a moving object
JP2020050517A (en) Congestion detection system for escalator
JP2020148674A (en) Vehicle detector, vehicle detection method, and program
JP2019166996A (en) Monitoring device and monitoring method
JP2005056441A (en) Traffic flow measurement device
JP2010086418A (en) Object discriminating device

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20101001

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20120713

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20120731

A521 Request for written amendment filed

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20120928

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: 20130305

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20130402

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20160412

Year of fee payment: 3

R150 Certificate of patent or registration of utility model

Free format text: JAPANESE INTERMEDIATE CODE: R150

Ref document number: 5241461

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

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

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250