JP2015178405A - In-car behavior detection device - Google Patents

In-car behavior detection device Download PDF

Info

Publication number
JP2015178405A
JP2015178405A JP2014056757A JP2014056757A JP2015178405A JP 2015178405 A JP2015178405 A JP 2015178405A JP 2014056757 A JP2014056757 A JP 2014056757A JP 2014056757 A JP2014056757 A JP 2014056757A JP 2015178405 A JP2015178405 A JP 2015178405A
Authority
JP
Japan
Prior art keywords
detection
image data
car
behavior
subject
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.)
Granted
Application number
JP2014056757A
Other languages
Japanese (ja)
Other versions
JP6096695B2 (en
Inventor
卓馬 吉村
Takuma Yoshimura
卓馬 吉村
拓也 國貞
Takuya Kunisada
拓也 國貞
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.)
Hitachi Building Systems Co Ltd
Original Assignee
Hitachi Building Systems 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 Hitachi Building Systems Co Ltd filed Critical Hitachi Building Systems Co Ltd
Priority to JP2014056757A priority Critical patent/JP6096695B2/en
Publication of JP2015178405A publication Critical patent/JP2015178405A/en
Application granted granted Critical
Publication of JP6096695B2 publication Critical patent/JP6096695B2/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Indicating And Signalling Devices For Elevators (AREA)
  • Maintenance And Inspection Apparatuses For Elevators (AREA)
  • Cage And Drive Apparatuses For Elevators (AREA)
  • Closed-Circuit Television Systems (AREA)
  • Alarm Systems (AREA)
  • Image Analysis (AREA)

Abstract

PROBLEM TO BE SOLVED: To provide an in-car behavior detection device which allows the detection of violent behavior or the detection of staying without any erroneous detection even when a car having a glass transparent part is used.SOLUTION: An in-car behavior detection device includes: a door closing detection section 7 for outputting a door closing detection signal when a door 3 of a car 1 is closed; an image input section 6 for extracting subject image data on the near side of a glass transparent part 2 from image data obtained by taking an image of the inside of the car 1 when receiving the door closing detection signal from the door closing detection section 7; a behavior image creation section 8 for creating difference image data from the subject image data within a prescribed time extracted by the image input section 6; a behavior detection section 9 for discriminating the detection of violent behavior and the detection of staying on the basis of a comparison between the subject image data and the difference image data; and a notification section 10 for making a notification when the behavior detection section 9 performs at least one of the detection of violent behavior and the detection of staying.

Description

本発明は、エレベータ乗りかご内の人物の暴れや倒れ等の異常行動を検出するための乗りかご内挙動検出装置に関する。   The present invention relates to an in-car behavior detection device for detecting abnormal behavior such as rampage or falling of a person in an elevator car.

エレベータ乗りかご内の人物の暴れや倒れ等を検出する乗りかご内挙動検出装置として、乗りかご内に設置された防犯カメラからの映像の1コマ1コマの差分を抽出し、一定の時間の差分が大きい場合は暴れ行動と判定し、乗りかご内の人物に対し自動で注意喚起を行ったり、一定の時間、乗りかご内の人物の動き成分が少ない場合は倒れ状態と判定し、ドアを開けたりブザーで乗りかご外に危険を伝えることが知られている。しかし、防犯カメラの付近に背の高い人物が居ると、背の高い人物によって防犯カメラの映像に死角が生まれ、死角の先の状態を検出できない他に、防犯カメラに近くなることで、多少の動きであっても大きな差分を抽出してしまうことから、挙動検出装置の検出漏れや誤検出に繋がってしまう。この問題を解決するために、乗りかご内に、複数の防犯カメラを、視野が重なるように設置し、これら複数の防犯カメラの映像から、任意の視点での仮想映像を生成し、防犯カメラの死角をなくしたかご内映像監視装置が提案されている(特許文献1参照)。   As an in-car behavior detection device that detects the rampage or fall of a person in an elevator car, the difference between the frames from the security camera installed in the car is extracted, and the difference of a certain time If it is large, it will be judged as rampage, and the person in the car will be alerted automatically, or if there is little movement component of the person in the car for a certain period of time, it will be judged as fallen and the door will be opened. A buzzer is known to convey danger outside the car. However, if there is a tall person in the vicinity of the security camera, a blind spot is created in the video of the security camera by the tall person, and in addition to being unable to detect the state behind the blind spot, Even if it is a motion, since a big difference is extracted, it will lead to a detection omission and a false detection of a behavior detection apparatus. In order to solve this problem, multiple security cameras are installed in the car so that their fields of view overlap, and virtual images from arbitrary viewpoints are generated from the images of these multiple security cameras. An in-car image monitoring device that eliminates blind spots has been proposed (see Patent Document 1).

特開2005−15094号公報JP-A-2005-15094

しかしながら、従来の乗りかご内挙動検出装置によれば、乗りかご内を任意の視点で監視して防犯カメラの死角をなくすことができるが、ドアにガラス窓が付いているエレベータ乗りかごや、乗りかごの側面や背面がガラス状となっている展望用エレベータ等では、ガラス透過部を通して外部の変化を動き成分として検出するため、外部の変化が大きい場合には異常と判断し、誤検出が発生してしまう。   However, according to the conventional behavior detection device in the car, the inside of the car can be monitored from any viewpoint and the blind spot of the security camera can be eliminated. However, the elevator car with the glass window on the door or the car In an elevator for observation that has a glass-like side or back, the external change is detected as a movement component through the glass transmission part, so if the external change is large, it is judged as abnormal and a false detection occurs. End up.

本発明の目的は、ガラス透過部を有した乗りかごを使用しても、誤検出をすることなく暴れ検知または滞留検知を行えるようにした乗りかご内挙動検出装置を提供することにある。   An object of the present invention is to provide an in-car behavior detection device that can perform a rampage detection or a stay detection without erroneous detection even when a car having a glass transmission part is used.

本発明は上記目的を達成するために、ガラス透過部を有するエレベータの乗りかご内にカメラを設置し、前記カメラの画像データを使用して乗客の暴れや滞留を検出する乗りかご内挙動検出装置において、前記乗りかごのドアが閉じたときにドア閉検出信号を出力するドア閉検出手段と、前記ドア閉検出手段によりドア閉検出信号を受けたとき前記乗りかご内を撮影した画像データからガラス透過部よりも手前側の被写体画像データを抽出する画像入力手段と、前記画像入力手段が抽出した所定時間内の前記被写体画像データから差分画像データを生成する挙動画像作成手段と、前記被写体画像データと前記差分画像データとの比較から暴れ検知と滞留検知とを判別する挙動検知手段と、前記挙動検知手段が暴れ検知または滞留検知の少なくとも一方を検知したときに報知する報知手段とを設けたことを特徴とする。   In order to achieve the above-mentioned object, the present invention provides a behavior detecting device in a car that installs a camera in a car of an elevator having a glass transmission part and detects passengers' rampage and stay using image data of the camera. And a door close detection means for outputting a door close detection signal when the door of the car is closed, and glass from image data obtained by photographing the inside of the car when the door close detection signal is received by the door close detection means. Image input means for extracting subject image data in front of the transmission unit, behavior image creation means for generating difference image data from the subject image data within a predetermined time extracted by the image input means, and the subject image data Behavior detection means for discriminating between rampage detection and stagnation detection from a comparison between the difference image data and the difference image data; Characterized in that also provided with informing means for informing when it detects one.

上述の構成によれば、ガラス透過部を有した乗りかごを使用しても、ガラス透過部よりも乗り場側における被写体や、景色の変化などの影響を容易に除くことができ、また乗りかごのドアが閉じた状態での画像データを使用して誤検出の要因を除去し、乗りかごの外部での大きな変化による誤検出を防止して精度の高い暴れ検知または滞留検知を行うことができる。   According to the above-described configuration, even if a car having a glass transmissive part is used, it is possible to easily remove the influence of a subject on the landing side and a change in scenery from the glass transmissive part. By using the image data in a state where the door is closed, the cause of erroneous detection can be removed, and erroneous detection due to a large change outside the car can be prevented, and high-precision rampage detection or stay detection can be performed.

また本発明は、上述の構成に加えて、前記カメラとしてステレオカメラを用いたことを特徴とする。   In addition to the above-described configuration, the present invention is characterized in that a stereo camera is used as the camera.

上記の構成によれば、一台のステレオカメラを使用した簡単な構成で、乗りかごのドアに設けたガラス透過部よりも乗り場側における被写体や、景色の変化などの影響を容易に除くことができ、また乗りかごのドアが閉じた状態での画像データを使用して誤検出の要因を除去し、乗りかごの外部での大きな変化による誤検出を防止して精度の高い暴れ検知または滞留検知を行うことができる。   According to the above configuration, it is possible to easily remove the influence of the subject on the landing side and the change of scenery from the glass transmission part provided on the door of the car with a simple configuration using one stereo camera. It is possible to use the image data when the car door is closed, eliminate the cause of false detection, and prevent false detection due to a large change outside the car, so that high-precision rampage detection or stay detection It can be performed.

本発明による乗りかご内挙動検出装置によれば、ガラス透過部を有する乗りかごにおいても、ガラス透過部よりも乗り場側における被写体や、景色の変化などの影響を容易に除くことができ、そうした外部での大きな変化による誤検出を防止して精度の高い暴れ検知または滞留検知を行うことができる。   According to the behavior detecting device in the car according to the present invention, even in a car having a glass transmission part, it is possible to easily eliminate the influence of the subject on the landing side and the change of the scenery from the glass transmission part. It is possible to prevent erroneous detection due to a large change in the temperature and to perform high-precision rampage detection or stay detection.

本発明の一実施の形態による乗りかご内挙動検出装置の概略構成図である。1 is a schematic configuration diagram of an in-car behavior detection device according to an embodiment of the present invention. 図1に示した乗りかご内挙動検出装置の動作を示すフローチャートである。It is a flowchart which shows operation | movement of the behavior detection apparatus in a car shown in FIG.

以下、本発明の実施の形態を図面に基づいて説明する。   Hereinafter, embodiments of the present invention will be described with reference to the drawings.

図1は、本発明の一実施の形態による乗りかご内挙動検出装置の概略構成図である。   FIG. 1 is a schematic configuration diagram of an in-car behavior detection device according to an embodiment of the present invention.

エレベータの乗りかご1には、ガラス窓などのガラス透過部2が付いたドア3が取り付けられており、乗りかご1内には防犯用のステレオカメラ4が設置されている。ここでステレオカメラ4は、ドア3を撮影する側の位置に配置されて乗りかご1内の撮影を行っている。図示を省略しているが、乗り場にも同様の乗り湯ドアが設置されている。   A door 3 with a glass transmission part 2 such as a glass window is attached to the elevator car 1, and a crime prevention stereo camera 4 is installed in the car 1. Here, the stereo camera 4 is arranged at a position on the side where the door 3 is photographed and performs photographing inside the car 1. Although not shown, a similar hot water door is also installed at the landing.

ステレオカメラ4により撮影した場合、画像データから予め設定した距離内の画像データのみを抽出してかご内画像データを作成することができる。この特徴を生かして、ステレオカメラ4は撮影した画像データからガラス透過部2よりも手前側の画像データを抽出して新たなかご内画像データを作成し、この新たなかご内画像データを挙動検出装置5の画像入力手段6に出力するようにしている。   When the image is taken by the stereo camera 4, only the image data within a preset distance can be extracted from the image data to create the in-car image data. Taking advantage of this feature, the stereo camera 4 extracts the image data on the near side of the glass transmission part 2 from the captured image data to create new in-car image data, and detects the behavior of the new in-car image data. The image is output to the image input means 6 of the apparatus 5.

画像入力手段6は、新たなかご内画像データを使用することになるため、ガラス透過部2よりもさらに遠方、つまりガラス透過部2よりも乗り場側における被写体や、外部の景色の変化の影響を除くことができる。また画像入力部6は、所定時間毎にステレオカメラ4から入力された新たなかご内画像データを取り込んでいるが、新たなかご内画像データのうちドア閉検出部7からドア閉検出信号が入力されているときのものだけを使用し、そこに被写体が存在するかどうかを判定し、被写体が存在する場合は、それを被写体画像データとして挙動画像作成部8に出力する。   Since the image input means 6 uses new in-car image data, the image input means 6 is more distant than the glass transmission part 2, that is, the influence of the subject on the landing side of the glass transmission part 2 and the change of the external scenery. Can be excluded. Further, the image input unit 6 takes in new in-car image data input from the stereo camera 4 every predetermined time, and the door close detection signal is input from the door close detection unit 7 in the new in-car image data. When the subject is present, it is determined whether or not the subject is present. If the subject is present, it is output to the behavior image creating unit 8 as subject image data.

挙動画像作成部8は、画像入力部6から受け取った複数の被写体画像データを使用して被写体の動きに対応する被写体差分画像データを作成する。このとき、被写体に動きがあれば被写体差分画像データは生成され、被写体に動きが無ければ被写体作動画像データは生成されない。被写体差分画像データが生成された場合、被写体画像データと被写体差分画像データとを挙動検知部9に出力する。   The behavior image creation unit 8 creates subject difference image data corresponding to the movement of the subject using the plurality of subject image data received from the image input unit 6. At this time, subject difference image data is generated if the subject moves, and subject action image data is not generated if the subject does not move. When the subject difference image data is generated, the subject image data and the subject difference image data are output to the behavior detection unit 9.

挙動検知部9は、挙動画像作成部8から入力された被写体画像データと被写体差分画像データとを比較して暴れに対応するかどうかを判定する。例えば、両データの大きさを比較し、被写体差分画像データの大きさが被写体画像データの大きさに対し、30%以上なら、暴れ検知と判定し報知部10に暴れ検知信号を出力する。   The behavior detection unit 9 compares the subject image data input from the behavior image creation unit 8 with the subject difference image data and determines whether to cope with the rampage. For example, the size of both data is compared, and if the size of the subject difference image data is 30% or more with respect to the size of the subject image data, it is determined that a rampage is detected, and a rampage detection signal is output to the notification unit 10.

一方、挙動検知部9は、被写体差分画像データに大きさがあるものの被写体差分画像データが所定時間を超えても余り被写体が動かない場合、滞留検知と判定する。例えば、10分を越えても、被写体差分画像データが所定の大きさが被写体画像データの大きさに対して5%未満の場合、乗りかご1から降車せずに被写体が移動せず、乗りかご内に滞留していると判断し、滞留検知と判定し報知部10に滞留検知信号を出力する。   On the other hand, when the subject difference image data has a size but the subject difference image data exceeds the predetermined time, the behavior detection unit 9 determines that the stay is detected. For example, even if it exceeds 10 minutes, if the subject difference image data has a predetermined size less than 5% of the size of the subject image data, the subject does not move from the car 1 without moving, and the car does not move. It is determined that it is staying inside, and it is determined that stay is detected, and a stay detection signal is output to the notification unit 10.

報知部10は、挙動検知部9から暴れ検知信号を受け取ったとき警報音を発生させ、図示しないエレベータの制御装置に信号を出力し、乗りかごを最寄り階に停止させドア3を開き、暴れている乗客の存在を報知することができる。一方、挙動検知部9から滞留検知信号を受信したとき報知部10は、警報音を発生させ、同様にエレベータを最寄り階に停止させてドア3を開き、滞留者の存在を報知することができる。   The notification unit 10 generates an alarm sound when receiving a ramp detection signal from the behavior detection unit 9, outputs a signal to an elevator control device (not shown), stops the car on the nearest floor, opens the door 3, and rampages. The presence of a passenger can be notified. On the other hand, when receiving the stay detection signal from the behavior detection unit 9, the notification unit 10 generates an alarm sound, similarly, stops the elevator on the nearest floor and opens the door 3 to notify the presence of the staying person. .

次に、図2に示したフローチャートを用いて、上述した挙動検出装置の処理動作を説明する。   Next, the processing operation of the behavior detection apparatus described above will be described using the flowchart shown in FIG.

挙動検出装置5の画像入力部6は、手順S1で、ドア閉検出部7からドア3のドア閉検出信号の有無を監視する。ドア閉検出信号の入力が無い場合、つまり乗りかご1のドア3が開いている状態では、手順S1の監視を継続する。一方、ドア閉検出信号の入力が有って乗りかご1のドア3が閉じられた場合、手順S2において、ステレオカメラ4からの新たな画像データを取り込み、ステレオカメラ4からガラス透過部2までの距離内の新たな画像データ内に被写体が存在しているかどうかを判定する。   The image input unit 6 of the behavior detection device 5 monitors the presence / absence of a door close detection signal of the door 3 from the door close detection unit 7 in step S1. When the door close detection signal is not input, that is, when the door 3 of the car 1 is open, the monitoring of the procedure S1 is continued. On the other hand, when the door close detection signal is input and the door 3 of the car 1 is closed, in step S2, new image data from the stereo camera 4 is taken in and the stereo camera 4 to the glass transmission part 2 are read. It is determined whether or not a subject exists in the new image data within the distance.

この判定の結果、新たな画像データ内に被写体が存在しなければ処理を終了する。しかし、手順S2での判定の結果、新たな画像データ内に被写体が存在すれば、被写体画像データとして挙動画像作成部8に出力する。これを受けた挙動画像作成部8は、手順3において、画像入力部6から入力された被写体画像データの差分画像である被写体差分画像データを作成し、被写体画像データと被写体差分画像データを挙動検知部9に出力する。   As a result of this determination, if there is no subject in the new image data, the process is terminated. However, if there is a subject in the new image data as a result of the determination in step S2, it is output to the behavior image creating unit 8 as subject image data. In response to this, the behavior image creation unit 8 creates subject difference image data that is a difference image of the subject image data input from the image input unit 6 in step 3, and detects behavior of the subject image data and the subject difference image data. To the unit 9.

手順S4において挙動検知部9は、挙動画像作成部8から入力された被写体画像データと被写体差分画像データを所定の閾値を用いて2値化して、2値化した画像部分の面積を比較し、その差の大きさが所定の割合、例えば30%以上の場合、暴れ検知と判定し報知部10に暴れ検知信号を出力する。   In step S4, the behavior detection unit 9 binarizes the subject image data and subject difference image data input from the behavior image creation unit 8 using a predetermined threshold, and compares the areas of the binarized image portions. When the magnitude of the difference is a predetermined ratio, for example, 30% or more, it is determined that the detection is a detection, and a detection detection signal is output to the notification unit 10.

暴れ検知信号を受信した報知部10は、手順S5において、暴れ検知の警報を鳴動し、エレベータの制御装置に信号を出力し処理を終了する。   In step S5, the notification unit 10 that has received the rampage detection signal sounds a rampage detection alarm, outputs a signal to the elevator control device, and ends the process.

これに対して挙動検知部9は、手順S4の判定で、被写体画像データと被写体差分画像データとの2値化した画像部分の面積を比較して所定の割合、例えば、30%未満の場合、手順S6に進む。手順S6において、2値化した画像部分の面積を比較し所定の割合、例えば5%未満かどうかを判定する。その結果、所定の割合、例えば5%以上の場合、手順1から繰返し実施する。しかし、2値化した画像部分の面積を比較し所定の割合、例えば、5%未満の場合、手順7でタイマーをカウントアップする。   On the other hand, the behavior detection unit 9 compares the areas of the binarized image portions of the subject image data and the subject difference image data in the determination of step S4, and when the ratio is less than a predetermined ratio, for example, less than 30%, Proceed to step S6. In step S6, the areas of the binarized image portions are compared to determine whether or not a predetermined ratio, for example, less than 5%. As a result, when the ratio is a predetermined ratio, for example, 5% or more, the procedure 1 is repeated. However, the areas of the binarized image portions are compared, and if it is less than a predetermined ratio, for example, less than 5%, the timer is counted up in step 7.

次に、挙動検知部9は、手順8において、タイマーが所定の時限をカウントしたかどうか判定する。例えば10分をカウントアップしたかどうかを判定し、所定の時間を経過していない場合、確認のために手順1から処理を繰返し実施する。これに対して、手順S8の判定の結果、所定の時間を経過していた場合、滞留検知と判定し報知部10に滞留検知信号を出力する。   Next, the behavior detection unit 9 determines in step 8 whether the timer has counted a predetermined time period. For example, it is determined whether or not 10 minutes have been counted up. If a predetermined time has not elapsed, the process is repeated from step 1 for confirmation. On the other hand, if a predetermined time has passed as a result of the determination in step S <b> 8, it is determined as stay detection and a stay detection signal is output to the notification unit 10.

滞留検知信号を受信した報知部10は、手順S9において、滞留検知の警報を鳴動し、エレベータの制御装置に信号を出力して処理を終了する。   In step S9, the notification unit 10 that has received the stay detection signal sounds a stay detection alarm, outputs a signal to the elevator control device, and ends the process.

尚、本発明は、挙動検知部9における暴れ検知または滞留検知として上述した実施の形態で説明した方法に限定するものではない。また、乗りかご1内に設置したカメラとしてステレオカメラ4を使用しているが、単眼の防犯カメラを複数台を乗りかごに設置する構成としても良い。   In addition, this invention is not limited to the method demonstrated in embodiment mentioned above as the rampage detection in the behavior detection part 9, or a stay detection. Moreover, although the stereo camera 4 is used as a camera installed in the car 1, a configuration may be adopted in which a plurality of monocular security cameras are installed in the car.

上述した本発明による乗りかご内挙動検出装置5は、ガラス透過部2を有したエレベータの乗りかご1内にカメラ4を設置し、カメラ4の画像データを使用して乗客の暴れや滞留を検出する乗りかご内挙動検出装置において、乗りかご1のドア3が閉じたときにドア閉検出信号を出力するドア閉検出部7と、ドア閉検出部7によりドア閉検出信号を受けたとき乗りかご1内を撮影した画像データからガラス透過部2よりも手前側の被写体画像データを抽出する画像入力部6と、画像入力部6が抽出した所定時間内の前記被写体画像データから差分画像データを生成する挙動画像作成部8と、被写体画像データと前記差分画像データとの比較から暴れ検知と滞留検知とを判別する挙動検知部9と、挙動検知部9が暴れ検知または滞留検知の少なくとも一方を検知したときに報知する報知部10とを設けたことを特徴とする。   The above-described behavior detecting device 5 in the car according to the present invention has the camera 4 installed in the elevator car 1 having the glass transmission part 2 and detects the passenger's rampage and stay using the image data of the camera 4. In the in-car behavior detecting apparatus, the door closing detection unit 7 that outputs a door closing detection signal when the door 3 of the car 1 is closed, and the car when the door closing detection unit 7 receives the door closing detection signal. An image input unit 6 for extracting subject image data on the near side of the glass transmission unit 2 from image data obtained by photographing the inside of the image 1 and differential image data are generated from the subject image data within a predetermined time extracted by the image input unit 6 The behavior image creating unit 8, the behavior detection unit 9 that discriminates between the detection of the rampage and the stay detection from the comparison between the subject image data and the difference image data, and the behavior detection unit 9 has little detection of the rampage or staying Characterized by providing a notification unit 10 which notifies when it detects one and also.

上述の構成によれば、ガラス透過部2を有した乗りかご1においても、ガラス透過部2よりも乗り場側における被写体や、景色の変化などの影響を容易に除くことができ、また乗りかご1のドア3が閉じた状態での画像データを使用して誤検出の要因を除去し、乗りかご1の外部での大きな変化による誤検出を防止して精度の高い暴れ検知または滞留検知を行うことができる。   According to the above-described configuration, even in the car 1 having the glass transmissive part 2, it is possible to easily remove the influence of the subject on the landing side and the change of the scenery from the glass transmissive part 2. The image data with the door 3 closed is used to eliminate the cause of false detection, and prevent false detection due to a large change outside the car 1 to perform high-precision rampage detection or stay detection. Can do.

また本発明は、上述の構成に加えて、前記カメラとしてステレオカメラ4を用いたことを特徴とする。   In addition to the above-described configuration, the present invention is characterized in that a stereo camera 4 is used as the camera.

上記の構成によれば、一台のステレオカメラ4を使用した簡単な構成で、乗りかご1のガラス透過部2よりも乗り場側における被写体や、景色の変化などの影響を容易に除くことができ、乗りかご1の外部での大きな変化による誤検出を防止して精度の高い暴れ検知または滞留検知を行うことができる。   According to the above configuration, it is possible to easily remove the influence of the subject on the landing side and the change of scenery from the glass transmission part 2 of the car 1 with a simple configuration using one stereo camera 4. Thus, it is possible to prevent erroneous detection due to a large change outside the car 1 and to perform high-precision rampage detection or stay detection.

1 乗りかご
2 ガラス透過部
3 ドア
4 ステレオカメラ
5 挙動検知装置
6 画像入力部
7 ドア閉検出部
8 挙動画像作成部
9 挙動検知部
10 報知部
DESCRIPTION OF SYMBOLS 1 Car 2 Glass transmission part 3 Door 4 Stereo camera 5 Behavior detection apparatus 6 Image input part 7 Door closing detection part 8 Behavior image creation part 9 Behavior detection part 10 Notification part

Claims (2)

ガラス透過部を有したエレベータの乗りかご内にカメラを設置し、前記カメラの画像データを使用して乗客の暴れや滞留を検出する乗りかご内挙動検出装置において、
前記乗りかごのドアが閉じたときにドア閉検出信号を出力するドア閉検出手段と、前記ドア閉検出手段によりドア閉検出信号を受けたとき前記乗りかご内を撮影した画像データから前記ガラス透過部よりも手前側の被写体画像データを抽出する画像入力手段と、前記画像入力手段が抽出した所定時間内の前記被写体画像データから差分画像データを生成する挙動画像作成手段と、前記被写体画像データと前記差分画像データとの比較から暴れ検知と滞留検知とを判別する挙動検知手段と、前記挙動検知手段が暴れ検知または滞留検知の少なくとも一方を検知したときに報知する報知手段とを設けたことを特徴とする乗りかご内挙動検出装置。
In a behavior detecting device in a car that installs a camera in a car of an elevator having a glass transmission part and detects passenger rampage and stay using image data of the camera,
A door closing detection means for outputting a door closing detection signal when the door of the car is closed; and glass transmission from image data obtained by photographing the inside of the car when the door closing detection signal is received by the door closing detection means. Image input means for extracting subject image data on the near side of the unit, behavior image creation means for generating difference image data from the subject image data within a predetermined time extracted by the image input means, the subject image data, Behavior detecting means for discriminating between rampage detection and stay detection from comparison with the difference image data, and notification means for notifying when the behavior detecting means detects at least one of rampage detection or stay detection are provided. Car behavior detection device.
前記カメラとしてステレオカメラを用いたことを特徴とする請求項1に記載の乗りかご内挙動検出装置。   The in-car behavior detection apparatus according to claim 1, wherein a stereo camera is used as the camera.
JP2014056757A 2014-03-19 2014-03-19 In-car behavior detector Active JP6096695B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2014056757A JP6096695B2 (en) 2014-03-19 2014-03-19 In-car behavior detector

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2014056757A JP6096695B2 (en) 2014-03-19 2014-03-19 In-car behavior detector

Publications (2)

Publication Number Publication Date
JP2015178405A true JP2015178405A (en) 2015-10-08
JP6096695B2 JP6096695B2 (en) 2017-03-15

Family

ID=54262758

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2014056757A Active JP6096695B2 (en) 2014-03-19 2014-03-19 In-car behavior detector

Country Status (1)

Country Link
JP (1) JP6096695B2 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107215733A (en) * 2017-05-25 2017-09-29 安徽味唯网络科技有限公司 A kind of method that elevator faults are sought help in time
CN107777498A (en) * 2017-11-20 2018-03-09 江苏省特种设备安全监督检验研究院 Act of violence detection method in a kind of lift car
JP2018144983A (en) * 2017-03-08 2018-09-20 株式会社日立ビルシステム Elevator, and diagnostic system and diagnostic method of in-car abnormality detecting device

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0997337A (en) * 1995-09-29 1997-04-08 Fuji Heavy Ind Ltd Trespasser monitor device
JP2000335854A (en) * 1999-05-24 2000-12-05 Hitachi Building Systems Co Ltd Crime prevention device for observation elevator
JP2008195495A (en) * 2007-02-14 2008-08-28 Hitachi Ltd Image monitoring device
JP2009035375A (en) * 2007-08-01 2009-02-19 Hitachi Building Systems Co Ltd Abnormality monitoring device in elevator car
JP2011063416A (en) * 2009-09-18 2011-03-31 Toshiba Elevator Co Ltd Violence detector and violence detecting method
JP2013056754A (en) * 2011-09-08 2013-03-28 East Japan Railway Co Door opening and closing determination device, device for monitoring state in car, door opening and closing determination method, and program
JP2013119461A (en) * 2011-12-07 2013-06-17 Mitsubishi Electric Building Techno Service Co Ltd Apparatus for monitoring inside of elevator car

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0997337A (en) * 1995-09-29 1997-04-08 Fuji Heavy Ind Ltd Trespasser monitor device
JP2000335854A (en) * 1999-05-24 2000-12-05 Hitachi Building Systems Co Ltd Crime prevention device for observation elevator
JP2008195495A (en) * 2007-02-14 2008-08-28 Hitachi Ltd Image monitoring device
JP2009035375A (en) * 2007-08-01 2009-02-19 Hitachi Building Systems Co Ltd Abnormality monitoring device in elevator car
JP2011063416A (en) * 2009-09-18 2011-03-31 Toshiba Elevator Co Ltd Violence detector and violence detecting method
JP2013056754A (en) * 2011-09-08 2013-03-28 East Japan Railway Co Door opening and closing determination device, device for monitoring state in car, door opening and closing determination method, and program
JP2013119461A (en) * 2011-12-07 2013-06-17 Mitsubishi Electric Building Techno Service Co Ltd Apparatus for monitoring inside of elevator car

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2018144983A (en) * 2017-03-08 2018-09-20 株式会社日立ビルシステム Elevator, and diagnostic system and diagnostic method of in-car abnormality detecting device
CN107215733A (en) * 2017-05-25 2017-09-29 安徽味唯网络科技有限公司 A kind of method that elevator faults are sought help in time
CN107777498A (en) * 2017-11-20 2018-03-09 江苏省特种设备安全监督检验研究院 Act of violence detection method in a kind of lift car

Also Published As

Publication number Publication date
JP6096695B2 (en) 2017-03-15

Similar Documents

Publication Publication Date Title
KR101485022B1 (en) Object tracking system for behavioral pattern analysis and method thereof
JP6242966B1 (en) Elevator control system
JP2009014712A (en) Object detecting device and gate device using the same
JP5833995B2 (en) Elevator abnormality monitoring device
JP6096695B2 (en) In-car behavior detector
EP3527525A1 (en) Elevator door control system for detection of obstacle trapping
JP6105105B2 (en) Disaster prevention equipment and obstacle detection device
TW201313597A (en) Safe control device and method for lift
TWI469098B (en) Security monitor system and method thereof
JP5090935B2 (en) Monitoring system and monitoring method
JP2007326380A (en) Security device and monitoring method
JP5170731B2 (en) Traffic monitoring system
TW201510945A (en) A method for identifying personnel and a security booth prevention system thereof
JP7021652B2 (en) In-car monitoring device
JP2014084064A (en) Platform door monitoring device and platform door monitoring method
WO2017104660A1 (en) Device for recording crime scene of shoplifting
JP5444103B2 (en) Reporting device
KR101629219B1 (en) An elevator monitor that restricts access through face recognition
CN110536854B (en) Operation prevention device for elevator
JP2012011988A (en) Detecting device of string-like foreign matter for railway vehicle door
KR101582454B1 (en) Method for cognition of movement object in photographing image and system for prevention of vehicle boarding accident
US11341773B2 (en) Detection device and control method of the same
JP7164395B2 (en) Intrusion detection systems, detectors and intrusion detection management systems
KR101223606B1 (en) System for intelligently monitoring image for monitoring elevator
JP6978986B2 (en) Warning system, warning control device and warning method

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20160222

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: 20170124

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20170126

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20170216

R150 Certificate of patent or registration of utility model

Ref document number: 6096695

Country of ref document: JP

Free format text: JAPANESE INTERMEDIATE CODE: R150