JP2007048249A - Safety monitoring system - Google Patents

Safety monitoring system Download PDF

Info

Publication number
JP2007048249A
JP2007048249A JP2005261293A JP2005261293A JP2007048249A JP 2007048249 A JP2007048249 A JP 2007048249A JP 2005261293 A JP2005261293 A JP 2005261293A JP 2005261293 A JP2005261293 A JP 2005261293A JP 2007048249 A JP2007048249 A JP 2007048249A
Authority
JP
Japan
Prior art keywords
video
monitoring
surveillance
monitoring system
intruder
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.)
Pending
Application number
JP2005261293A
Other languages
Japanese (ja)
Inventor
Yasuyuki Murai
保之 村井
Kiyohiko Nakazawa
齋彦 中澤
Ryuji Ohashi
竜二 大橋
Michitsugu Iwakubo
命嗣 岩久保
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.)
ARA SOFTWARE KK
Original Assignee
ARA SOFTWARE KK
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 ARA SOFTWARE KK filed Critical ARA SOFTWARE KK
Priority to JP2005261293A priority Critical patent/JP2007048249A/en
Publication of JP2007048249A publication Critical patent/JP2007048249A/en
Pending legal-status Critical Current

Links

Abstract

<P>PROBLEM TO BE SOLVED: To provide a picture analysis system capable of extracting dangerous events or the like for reducing a load of a watcher, wherein danger or the like is mainly determined by a watcher watching pictures, in regard to a safety monitoring system using picture information to carry out monitoring. <P>SOLUTION: An intelligent picture monitoring system is composed by using an intelligent picture analysis system capable of carrying out analysis in units of element of a monitoring object in a picture. An analysis function of the analysis system is used to monitor activities of the monitoring object. By having the intelligent picture monitoring system carry out judgement of monitoring for danger prevention carried out by the watcher, and transmission work to related departments when there is a dangerous situation, the load of the watcher can be reduced. At the same time, a movement record of a person, an automobile, or the like becoming the monitoring object such as an intruder, namely, an activity locus is displayed as a locus in the picture, and a measure to crime prevention is also taken by knowing the intrusion point. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

発明が属する技術分野Technical field to which the invention belongs

本発明は映像による安全監視システムに関する  The present invention relates to a video safety monitoring system.

従来の映像による安全監視システムは〔図1〕に示す通り、監視地点にある1台又は複数の監視カメラ101で得られた映像を有線または無線で映像伝送装置102を使ってて監視室108に送信し、該監視室においてその送られて来た映像をシステム統合サーバ103が記録装置104、表示装置105、プリンター106等に配信し、監視カメラ101から送信された映像情報を該表示装置等を介して監視員107が監視し、監視地点で発生した危険な状態もしくは異常な状態等監視項目、例えば不審者の軌道内侵入、踏切が遮断されている踏切内に人物、物体等の存在等であるが、を監視員が発見し、関係者もしくは関係部署に危険もしくは警告を伝達し同時にその映像は常時記録し続けるものであった〔特許文献1〕。該安全監視システムでは監視員が該映像から目を離さず注視し続けて監視する必要がる。  As shown in FIG. 1, a conventional video safety monitoring system uses a video transmission device 102 to connect a video obtained by one or a plurality of surveillance cameras 101 at a monitoring point to a monitoring room 108. In the monitoring room, the system integration server 103 distributes the transmitted video to the recording device 104, the display device 105, the printer 106, and the like, and the video information transmitted from the monitoring camera 101 is transmitted to the display device and the like. Monitoring items are monitored through monitoring items such as dangerous or abnormal conditions occurring at the monitoring point, such as intrusion of suspicious persons in the trajectory, presence of persons, objects, etc. in the level crossing where the level crossing is blocked However, a supervisor discovered the situation and transmitted a danger or warning to a related person or a related department, and at the same time, the video was constantly recorded [Patent Document 1]. In the safety monitoring system, it is necessary for a monitoring person to keep an eye on the video and keep monitoring.

さらに、複数の監視地点の映像情報が集められる監視室108においては、該監視員107が前記複数の監視地点のカメラからの映像を一度に監視できるように複数の表示装置をならべて、または単一の表示装置の画面を分割〔特許文献2〕するなどして、全ての監視点の映像を常時表示させて監視したり、もしくは単一または複数の表示装置に複数の監視点の映像を一定の短い時間ごとに順次切り替えて表示する方法によって監視が実施されており、前記監視員107はそれらの表示装置の映像を見て監視を続け、監視項目すなわち不審者の軌道内侵入や踏切が遮断されている踏切内に人物、物体等の存在等の異常、危険等の監視項目にあたる事象を見つけ出し、あらかじめ決められた場所、人達に伝達する。また記録装置では常時映像を記録し続ける。  Further, in the monitoring room 108 where the video information of a plurality of monitoring points is collected, the monitor 107 arranges a plurality of display devices so that the images from the cameras at the plurality of monitoring points can be monitored at a time, or simply. Divide the screen of one display device [Patent Document 2] to monitor all the monitoring point images at all times or to monitor a plurality of monitoring point images on a single or multiple display devices Monitoring is carried out by a method of sequentially switching and displaying every short period of time, and the monitor 107 continues to monitor by watching the video of those display devices, and the monitoring items, that is, the suspicious person intrusion in the orbit and crossing are blocked. An event corresponding to a monitoring item such as an abnormality such as the presence of a person, an object, or a danger in a level crossing is detected and transmitted to a predetermined place and people. In addition, the recording device continuously records video.

この監視員107の機能すなわち表示装置105の映像を見て異常や危険等を判定しそして通報する機能を映像解析システム等を使った監視システムに置換えようとする試みも行われている。(a)モーションディテクター検知システムで映像のフレーム間の変化を検出して判定する方式、(b)映像の変化領域の面積と移動速度から異常を検出するシステム(c)映像のピクセル集合のベクトルを算出することによって検出して判定する方法、即ちオプティカルフロー検知システムなどが良く知られている。  Attempts have also been made to replace the function of the monitoring person 107, that is, the function of judging an abnormality or danger by looking at the image of the display device 105 and notifying it with a monitoring system using an image analysis system or the like. (A) A method for detecting and determining a change between frames of a video by a motion detector detection system, (b) a system for detecting an abnormality from the area and moving speed of a video change area, and (c) a vector of a pixel set of a video A method of detecting and determining by calculating, that is, an optical flow detection system is well known.

前記(a)は現在最も一般化している検知システムであるが、映像中のどのような物体の変化でも同じように検出する、すなわち映像物体の特定、識別はできない。(b)は映像の変化領域の連続性と縦横比などから人物、自動車などの識別をして検出するため映像中の他の動くものたとえば草木等のゆれるものとの区別が出来る。しかし重なり合った複数の人物や人物の特定な行動、例えば車両軌道内の監視範囲にどこから侵入しその人物等の動きの軌跡を知ることはできない。(c)については最近開発された方法で土砂崩れ、河川の流量を計測する監視システムに利用されているが、監視カメラ自体が揺れたり映像内で動き廻る物体例えばカメラのレンズに滴が這うとか虫が這う等が有るときには誤動作が予想される。
特許公開2003−046989 特許公表2004−522379 特許公開2004−94518 ウエイン ウォルフ、バラック オザー、ティーハン リュー“スマートカメラを付加したシステム”電気電子学会コンピュータ部会 35(9)号 2002 Wayne Wolf、Burak Ozer,Tiehan Lv,“Smart Cameras as embedded system,”IEEE Computer,35(9)September 2002,
Although (a) is the most generalized detection system at present, any change in an object in a video is detected in the same way, that is, the video object cannot be identified or identified. (B) can be distinguished from other moving objects in the image such as vegetation because it identifies and detects a person, car, etc. from the continuity and aspect ratio of the change area of the image. However, it is impossible to know a plurality of overlapping persons or specific actions of the persons, for example, where to enter the monitoring range in the vehicle trajectory from where the person or the like moves. Regarding (c), a recently developed method has been used in a monitoring system that measures landslides and river flow, but the surveillance camera itself shakes or moves around in the image, for example, drops on the camera lens or insects. When there is a crawl, malfunction is expected.
Patent Publication 2003-046989 Patent publication 2004-522379 Patent Publication 2004-94518 Wayne Wolf, Barrack Ozer, Teehan Liu “System with Smart Cameras” Electrical and Electronic Society of Japan, Computer Group 35 (9) 2002 Wayne Wolf, Burak Ozer, Tiehan Lv, “Smart Cameras as embedded system,” IEEE 35, IEEE 35 ) September 2002

発明が解決しようとする課題Problems to be solved by the invention

通行の安全を確保し事故を最小限にするために設置されている例えば車両軌道の安全監視システムは事故の起こる可能性のある事象(ここでは監視内容と言う)を監視員が捉えて関係部署に通報し事故等を未然に防ぐことのために活用されるのが目的である。そのためには監視員の行う安全即ち危険防止のための判断とその情報を伝達するための処理の迅速化が求められる。監視員即ち人間の判断を待って伝達されるので時間的にはその判断の時間、警報の通達行為の時間がかかり、事前に知りえた危険情報の通知が時間を取ったために、事故回避と言う目的に十分に機能しないことも考えられる。その理由は、1つは表示装置の監視地点の映像で監視員が事象発生を見てから考え危険と判断し通報することによって時間がかかる問題がある。また危険の発生を即時発見するには監視のために該映像の注視を続けること即ち該映像から目を離さないことが必須であり、即ち監視員は該映像から目を離せない状態にあるはずである。さらに複数の表示装置を監視する場合にはその負荷の多さによる監視漏れ、監視判断の遅れは避けられない。つまり危険判断の即時対応は不可能に近い。2つめは通報する手段にかかわる。危険事象発生の予測を認知してから通報する作業にかかる時間差が事故への決定的な要因になることもあり、関係部署への連絡は確実に迅速にしなければならないと共に本当に生の情報を必要とする車両等の運転手、運転管理者に伝えなければ事故回避の本当の対策には近づかないことは周知のことである。3つ目は危険な行為をとる人物や物体に対し直接警告を伝えることも必要である。事故、災害の被害を最小限に抑えるためには、監視員だけに事故回避を期待するのは以上により大変難しいが、この監視員に対する過負荷状態に対し負荷軽減を図ることは欠かせない。同時にこの過負荷を低減することにより監視員の数すなわち人件費の軽減を図れると共に、危険等の発生を早くから即ち事前情報も含め情報収集し解析してその解析結果を関係する部署、人々にすばやく伝達できる監視システムが必要であり、各部署ではこの早い情報によって事故や災害にすばやく対策、行動を起こすことができるようになる。例えば車両の安全な運行にとって1つめの早期な危険検知と2つ目のすばやい危険情報の伝達が最も重要である。その効果をさらに出すためにはその近くを運行する列車の運転手への映像情報のすばやい伝達であり、運行管理室の人々への伝達であることも考えられる。これに対し色々な方法の監視システムの開発〔0004〕が試みられて来たが、判定を頼りにするにはまだ欠点もあり満足できていない。従って1つ目の監視員の負荷軽減にまで至っていない。  For example, in a vehicle track safety monitoring system that is installed to ensure traffic safety and minimize accidents, the monitoring staff grasps events that may cause accidents (herein referred to as monitoring contents) and related departments. It is intended to be used to prevent accidents from occurring. For this purpose, it is necessary to make a judgment for safety, that is, prevention of danger, and to speed up a process for transmitting the information. Since it is transmitted after waiting for the judgment of the observer, that is, human beings, it takes time for the judgment and the notification action of the alarm, and it takes time to notify the danger information that was known in advance, so it is said that the accident is avoided. It is also possible that it does not function well for the purpose. One of the reasons is that it takes time because it is considered dangerous after the observer sees the occurrence of the event with the video of the monitoring point of the display device and reports it as dangerous. In order to immediately detect the occurrence of danger, it is essential to keep an eye on the video for monitoring, that is, to keep an eye on the video, that is, the observer should be able to keep an eye on the video. It is. Further, when monitoring a plurality of display devices, monitoring omission due to the large load and delay in monitoring judgment are inevitable. In other words, immediate response to risk judgment is almost impossible. The second involves the means of reporting. The time difference between the work of recognizing the prediction of the occurrence of a dangerous event and the reporting may be a decisive factor for the accident, so it is necessary to make sure that the relevant departments are contacted promptly and that they really need raw information. It is well known that the real measures for avoiding accidents will not be approached unless it is communicated to the driver of the vehicle and the operation manager. Third, it is also necessary to convey a warning directly to a person or object who takes a dangerous action. In order to minimize the damage caused by accidents and disasters, it is much more difficult to expect only the observers to avoid the accident. However, it is essential to reduce the load against the overload state of the observers. At the same time, by reducing this overload, it is possible to reduce the number of observers, that is, labor costs, and to collect and analyze the occurrence of danger etc. early, that is, including prior information, and quickly analyze the analysis results to related departments and people. It is necessary to have a monitoring system that can be communicated, and each department can quickly take measures and take actions against accidents and disasters with this early information. For example, the first early detection of danger and the second quick transmission of danger information are most important for safe operation of the vehicle. In order to bring out the effect further, it is possible to transmit video information quickly to the driver of a train that operates in the vicinity, and to transmit it to people in the operation control room. On the other hand, the development of various monitoring systems [0004] has been attempted, but it is still unsatisfactory to rely on the judgment because of its drawbacks. Therefore, the load on the first observer has not been reduced.

課題を解決する為の手段Means to solve the problem

上記課題を解決するために、知的映像解析システム(〔非特許文献1〕に示すスマートカメラシステムの機能)を使うことにより可能となる。該知的映像解析システムの機能は図8に示す通り監視対象の部分部分の動きを独立に監視できることである。従来は画面内で動くものをひとつの形の塊りとして人物を捉えテンプレート化して追尾していたが〔特許文献3〕人物の頭、胴体、足や手などの要素部分に区別して動きを見ることはできなかった。即ち該解析システムによれば人物であること、ここでは物体でもおなじであるが、要素部分に区分して確認でき従って映像の中から人物のもっとも識別しやすい部分を選別でき、頭部であれば頭部を追跡のターゲットとすることで監視カメラが複数の人間を監視、追跡できる特長がある。監視員の判断行為即ち監視員が記憶している監視内容に照らしあわせて行う危険な事象の発生の判断、例えば踏切遮断にも関わらず踏切内に人物もしくは自動車等の物体が一定時間以上存在するとか、また踏切通路内から軌道内に踏み込み又は乗り入れて一定時間以上存在する等の危険行為等監視内容に該当する事象の発生の有無を判断する機能とその判断に基づいて警報を発信する機能を該知的映像監視システムが監視員に代わって行うことによって課題の解決が可能となる。
即ち監視カメラからの映像を該知的映像解析システムによって解析させ、危険等に該当すると判断する機能と、監視員や関係部署に危険や異常等の発生の情報を危険発生時の映像だけでなく危険と判断した時点の前の任意の時間からの映像やその映像の中に捕らえた侵入者もしくは危険物体の動きの軌跡を線で示した画像を重ねて表示した映像を監視員やあらかじめ決められた部署に送信し「危険表示」や「警報表示」及び音声で危険等を知らせる等危険情報を伝える機能を持つ知的映像監視システムを備える安全監視システムにより解決できる。
In order to solve the above problem, it is possible to use an intelligent video analysis system (function of the smart camera system shown in [Non-Patent Document 1]). The function of the intelligent video analysis system is to be able to independently monitor the movement of the part to be monitored as shown in FIG. In the past, a person was captured as a lump of shapes in the form of a moving object on the screen and templated and tracked. [Patent Document 3] Viewing movement by distinguishing between elements such as the person's head, torso, legs and hands. I couldn't. That is, according to the analysis system, it is the same as the person, here the object, but it can be confirmed by dividing it into element parts, so that the most easily distinguishable part of the person can be selected from the image. By using the head as the tracking target, the surveillance camera can monitor and track multiple people. The judgment action of the observer, that is, the judgment of the occurrence of a dangerous event performed in light of the monitoring contents stored by the observer, for example, there is an object such as a person or a car in the railway crossing for a certain time or more despite the crossing of the railway crossing In addition, a function that determines whether or not an event that falls under the monitoring content, such as a dangerous action such as being present for a certain period of time after entering or entering a track from a railroad crossing passage, and a function that issues an alarm based on the determination The intelligent video monitoring system can solve the problem by performing it on behalf of the supervisor.
In other words, the video from the surveillance camera is analyzed by the intelligent video analysis system, and it is judged that it falls under danger, etc., and information about the occurrence of danger and abnormality is sent to the surveillance staff and related departments as well as the video at the time of danger occurrence. An image from an arbitrary time before the point when it was determined to be dangerous, or a video in which a trajectory of an intruder or dangerous object captured in the image is superimposed and displayed as a line, can be monitored by a supervisor or in advance. This can be solved by a safety monitoring system equipped with an intelligent video monitoring system having a function of transmitting danger information, such as “danger indication” and “alarm indication” transmitted to a department, and notifying danger by voice.

発明の効果The invention's effect

本発明は次のような効果をもたらす。監視員の監視する機能即ち監視映像を見て前記危険等即ち監視内容を判定する機能とその危険であることの情報を伝達する機能を該映像解析システムを含む監視システム即ち知的映像監視システムに該機能を持たせることにより、監視員である人間による判断とその判断によって関係部署へ通知する行為にかかる時間を該監視システムによって行わしめることにより短縮すことができ、その分、危険を回避する行動を取るための時間が多くできる効果がある。危険防止には通報にかかる時間を短くすることは重要な事である。同時に監視員が常時監視映像を注視し続けなければならない負荷が大きく軽減される。即ち監視員の疲労が低減できると同時により少ない監視員でより多くの地点の監視が可能ともなる。また本発明の特徴である対象物の移動の軌跡を表示する線を描く画像を重ねて表示することにより監視対象物がどのように行動して危険行為に移行したか例えば踏切通路等のどこから車両軌道内に侵入したかとか、どこまで侵入して戻ったか、侵入時間とその軌跡を知ることで追跡監視が可能になり更なる防犯への対策を取れる等の効果がある。さらに危険を関係部署へ通知する手段を常時接続のインターネットまたはイントラネットなどのデジタル通信手段を使って行うことにより瞬時に危険情報である該映像と警報と追跡画像を同時に監視員、関係部署、とりわけ運転手や運行管理者である直接危険回避の行動をとる人々へも送ることができるシステムを構築できる。  The present invention brings about the following effects. A monitoring system that includes a video analysis system, that is, a function that determines the danger, that is, the content of the monitoring, and a function that transmits information about the danger are provided to the monitoring system that includes the video analysis system. By providing this function, it is possible to reduce the time taken by the person who is a supervisor and the act of notifying the related department by the judgment by the monitoring system, thereby avoiding danger accordingly. This has the effect of increasing the time taken to take action. In order to prevent danger, it is important to shorten the time required for reporting. At the same time, the load that the surveillance staff must keep watching the surveillance video at all times is greatly reduced. That is, the fatigue of the observer can be reduced, and at the same time, more points can be monitored with fewer observers. In addition, it is indicated how the monitored object behaves and shifts to a dangerous act by displaying an image that draws a line that displays the trajectory of the object that is a feature of the present invention. Knowing whether or not it has entered the orbit, how far it has entered, and the intrusion time and its trajectory makes it possible to follow up and monitor and take measures against further crime prevention. Furthermore, by using a digital communication means such as the Internet or an intranet that is always connected to notify the relevant department of the danger, the video, alarm, and tracking image, which are dangerous information, can be instantaneously displayed at the same time by the supervisor, the relevant department, especially the operation. It is possible to build a system that can be sent to people who take direct risk-avoidance actions such as hands and operation managers.

さらに従来の監視システムは監視カメラからの映像情報を常時記録するため大容量を必要とした。しかし前記映像解析システムによる判定機能を使って危険等を感知した時点の一定時間前の映像から記録を行うこともでき、その場合は必要とする記録容量は非常に少なくて済む経済効果が期待できる。さらに記録映像が限られるので、監視カメラによるプライバシーの侵害にもほとんど抵触せずに済む効果が期待できる。  Furthermore, the conventional monitoring system requires a large capacity because the video information from the monitoring camera is always recorded. However, it is also possible to record from the video a certain time before the time when danger is detected using the judgment function by the video analysis system, and in that case, the economic effect that only a very small recording capacity is required can be expected. . Furthermore, since the recorded video is limited, it can be expected that there will be almost no conflict with privacy infringement by the surveillance camera.

さらに本発明の効果を引き出すためには、危険等の判定機能即ち知的映像解析システムを含むコンピュータシステムを監視地点にある監視カメラ側に付属させることによってカメラ側の監視点で監視の判断機能を持つように構成し、危険と判断された場合には映像と共にその危険と判断した対象物や人物の行動の軌跡を表す線図等と共に危険を知らせる表示例えば赤字で危険の文字を表示するとかフラッシングするなどの表示と共に映像に重ねてもしくは映像の脇に表示させた映像の複合体を監視室やあらかじめ決められた部署に送信するが、監視カメラ側から監視室に該情報を送信する手段を従来の有線、無線による場合には監視カメラの数が多くなると通信回線の容量を大きくし設備費が嵩んだが危険と判断された時にその判断をしたカメラ側からのみ送信されるシステムであれば通信回線は小規模で済む経済効果がある。さらに他の通信手段として、監視カメラ側に付属する知的映像解析システムを含むコンピュータシステムの中にインターネット通信機能を持たせインターネット回線の常時接続によって該映像の複合体を監視する側に送り続けることもできる。特にこのインターネットによる情報の伝達は運転者や運行管理者の所にあるコンピュータの表示器にも映像情報等危険を知らせることも可能であり、危険情報が危険回避行動を取る直接の部署に即通知することができる効果を持つ。当然ではあるが監視室等情報を受け取る側からはいつでも希望の該監視カメラの映像を表示させて監視もできる。  Further, in order to bring out the effect of the present invention, a judgment function of danger or the like, that is, a judgment function of monitoring at a monitoring point on the camera side is provided by attaching a computer system including an intelligent video analysis system to the monitoring camera side at the monitoring point. If it is determined that there is a danger, the display will indicate the danger together with an image and a diagram representing the trajectory of the subject or person's action that is judged to be dangerous. A composite of a video that is superimposed on the video or displayed beside the video is transmitted to the monitoring room or a predetermined department, and means for transmitting the information from the monitoring camera side to the monitoring room is conventionally provided. If the number of surveillance cameras increases, the capacity of the communication line increases and the equipment costs increase. Communication If a system transmitted only La side line is small, requires economic effects. Further, as another communication means, a computer system including an intelligent video analysis system attached to the surveillance camera side has an Internet communication function, and continues to send the video complex to the side that monitors the video complex by always connecting the Internet line. You can also. In particular, the transmission of information via the Internet can also inform the computer display at the driver's or operation manager's computer display of danger such as video information, and the danger information is immediately notified to the direct department taking the risk avoidance action. Has an effect that can be. Of course, from the side receiving the information such as the monitoring room, the video of the desired monitoring camera can be displayed at any time for monitoring.

〔図1〕に従来型の映像による安全監視システムを示す。監視カメラであるカメラ101から監視のための映像が映像等送信装置102によって監視室108に送信され、監視室のシステム統合サーバ103で受信した該監視映像等を表示装置105および記録装置104、さらにプリンター106に配信する。表示そうち105に表示された該監視映像を監視員107が監視し、危険を発見した時にはアラームを鳴らすなど警報装置109によって記載されていないが関係者もしくは関係部署に伝達する。〔図2〕は〔図1〕の従来型映像による安全監視システムに知的映像解析システムを組み込ませたコンピュータシステム210を映像等送信装置202とシステム統合サーバ203の回線の途中に接続することによって、該機能の追加型である知的映像監視システムを構成したシステム図である。該コンピュータシステム210により監視カメラ201からの映像の中の映像対象物の要素部分例えば頭の部分や胴体の部分など要素部分を個別に監視することができる。〔図8〕はこの機能を具体的に図化して説明している。人物を例にして頭部801と胴体部802および手の部分802を分けてその動きを解析し右の図(b)は次の映像の画面において頭部811、胴体部813及び手の部分812の位置を解析した模式図である。下部の(c)、(d)の図は従来から行われている物体の画像認識による認識方法の模式図である。映像の中に捉えた人物の体全体もしくは自動車等の全体を監視対象としその動き即ち映像の次の画面での移動した人物の捉え方を示している。さらに〔図2〕は列車運転手等207に直接映像情報等を送信できることを示しており、通信は有線、無線だけでなく、インターネットによっても可能である。〔図3〕は本発明の基本をなす監視区域を指定した場合の例を示す図である。〔図3〕では道路302に続く踏切バーからの踏切の現場を示している。ここでは太い点線で囲まれた部分303が踏切内監視領域である。もちろん車両軌道304が本監視領域内を横切っている。このようにして監視領域を設定できることを示している。〔図4〕は知的映像解析システムの機能を使って、この設定された領域外もしくは領域内即ち設定をどちらにするかで監視内容がかわるが、例えばこの設定した範囲から軌道内に踏み出た単数もしくは複数の移動物体、ここでは侵入者401を監視し、監視領域303から侵入禁止領域に入ってからの行動の軌跡を線で表して映像に重ねて表示すると共に監視対象になった人物の任意の時間前からの軌跡も表示もできることを示している。例えば、足を踏み外して車両軌道内に踏み込んでも任意に設定された時間(例として5秒間)の範囲内であれば危険の情報は出さず、その設定時間以上の時間を軌道内に侵入したままの状態の場合は侵入者401と判断し危険情報を出す。FIG. 1 shows a conventional safety monitoring system using video. A video for monitoring is transmitted from the camera 101 which is a monitoring camera to the monitoring room 108 by the video transmission device 102, and the monitoring video received by the system integration server 103 in the monitoring room is displayed on the display device 105 and the recording device 104. Delivered to the printer 106. The monitoring image 107 displayed on the display edge 105 is monitored by the monitoring person 107, and when a danger is found, an alarm is sounded or the like, but it is not described by the alarm device 109 but is transmitted to the related person or related department. [FIG. 2] is obtained by connecting a computer system 210 in which an intelligent video analysis system is incorporated into the conventional video safety monitoring system of FIG. 1 in the middle of the line between the video transmission device 202 and the system integration server 203. FIG. 2 is a system diagram of an intelligent video surveillance system that is an additional type of the function. The computer system 210 can individually monitor the element parts of the video object in the video from the monitoring camera 201 such as the head part and the body part. FIG. 8 specifically illustrates this function. Taking a person as an example, the head 801, the torso 802, and the hand portion 802 are separated and their movements are analyzed. The right figure (b) shows the head 811, the torso portion 813, and the hand portion 812 on the screen of the next video. It is the schematic diagram which analyzed the position. The lower diagrams (c) and (d) are schematic diagrams of a conventional recognition method based on object image recognition. The whole body of a person captured in the video or the whole of a car or the like is monitored, and its movement, that is, how to recognize the moved person on the next screen of the video is shown. Further, [FIG. 2] shows that video information or the like can be directly transmitted to the train driver or the like 207, and communication is possible not only by wire and wireless but also by the Internet. FIG. 3 is a diagram showing an example in the case where a monitoring area that forms the basis of the present invention is designated. FIG. 3 shows a crossing site from a crossing bar following the road 302. Here, a portion 303 surrounded by a thick dotted line is a monitoring area in a crossing. Of course, the vehicle track 304 crosses the monitoring area. This shows that the monitoring area can be set. [FIG. 4] uses the function of the intelligent video analysis system to change the monitoring contents depending on whether the setting is outside or within the set area. One or more moving objects, in this case, the intruder 401 is monitored, the action trajectory after entering the intrusion prohibited area from the monitoring area 303 is displayed as a line and superimposed on the video, and the person who is the monitoring target It is shown that the trajectory from any time before can be displayed. For example, even if the foot is stepped off and stepped into the vehicle trajectory, if it is within the range of an arbitrarily set time (for example, 5 seconds), no danger information is output, and the time longer than the set time remains intruding into the track In the case of this state, it is determined that the person is an intruder 401 and danger information is output.

監視カメラの映像について、〔図5〕では侵入者503が持っていた物体を軌道内に放置502して離れたことを重大な危険と判断し、さらに、その侵入者503がその場所を立ち去るか戻って来た場合には不法な行為、すなわち不法な物品投棄者もしくは線路妨害者と判断して、その侵入者の移動の軌跡501を記録し続け線で表し、該安全監視システムは重大な危険情報として該監視映像に重ねて、もしくは表示装置の一部に最も目立つ方法表示し例えば警報音声と表示のフラッシング等も合わせて行わしめ、監視側即ち監視員や監視に関係する人達に警報等を出し続けることにより事故等の回避への情報を送り続ける。  Regarding the image of the surveillance camera, in FIG. 5, it is determined that it is a serious danger that the object held by the intruder 503 is left in the trajectory 502, and the intruder 503 leaves the place. When returning, it is judged as an illegal act, that is, an illegal dumper or a line jammer, and the trajectory 501 of the intruder is continuously recorded and represented by a line. Overlay the monitoring video as information, or display the most prominent method on a part of the display device, for example, alarm warning and display flushing etc. are also performed, and alarms etc. are given to the monitoring side, that is, monitoring personnel and people related to monitoring Continue to send information to avoid accidents and so on.

〔図6〕は知的映像監視システムを移植した安全監視システムで侵入禁止である監視領域に侵入した複数の物体についての監視につき説明をした図である。〔図6〕はすでに閉められた踏切の中にまだ監視対象物、図の例では二人の人間601,602もしくは自動車やバイク、自転車に乗った人物でも監視対象物となりうるが、これらが設定された時間の範囲を越えて存在していると判断した場合には関係部署に警報を送信するとともに危険状態にある人物もしくは自動車等にたいし〔図6〕には記されていないが音声等で危険情報を出すことも考えられる。踏切が閉じられたという情報は通常の踏み切りのアラームを鳴らす電気信号を参照しても良い。もしくは〔図6〕に示されているように画像解析技術を使って遮断機が下りたことを踏切バー603即ち遮断のための棒の位置によって判断し、該システムの監視機能をスタートさせても良い。[FIG. 6] is a diagram for explaining monitoring of a plurality of objects that have entered a monitoring area where entry is prohibited by a safety monitoring system in which an intelligent video monitoring system is transplanted. [FIG. 6] is an object to be monitored in a railroad crossing that has already been closed, and in the example shown in the figure, two people 601 and 602 or a person on a car, a motorcycle, or a bicycle can also be a monitoring object. If it is determined that it exists outside the specified time range, an alarm is sent to the relevant department, and it is not shown in FIG. It is also possible to issue danger information at The information that the level crossing is closed may refer to an electrical signal that sounds a normal level crossing alarm. Alternatively, as shown in FIG. 6, it is possible to use the image analysis technique to determine that the breaker has been lowered based on the position of the crossing bar 603, that is, the blocking bar, and start the monitoring function of the system. good.

〔図7〕は知的映像解析システムを移植したコンピュータシステムを監視地点にある各監視カメラ701に付属させ、あらかじめ設定された危険と判断する行動や動作を監視内容として該コンピュータシステムに記憶させておき監視カメラからの映像の監視対象物、例えば人物とか自動車、自転車等で代表される物体であるが、これらを映像解析して監視内容と比較し警報を出す条件を満たした時には警報の表示とその判定を下した時点より前の任意の時間からの映像を常時接続のインターネット回線702を介して監視室711やその他必要な部署または人々に該映像と危険情報および追跡した軌跡の線等を送信する。警報情報を含む該映像をシステム統合サーバ705に送信し映像表示装置707に表示させ、監視員710に目立つように危険表示信号例えば赤色で大きく表示させたり、フラッシングさせたりして危険等知らせる。同時に該映像の記録は映像記録装置708にて記録し、該警報情報はまた自動警報出力装置706から所定の宛先へ送られる。プリンター709は必要により関係画像を印刷する。インターネットによる情報のやりとりはシステム統合サーバを通すまでも無くインターネットの受信機を持つ部署例えば近くを運転中の列車の運転室等パソコンのあるどこにでも送ることができる。監視カメラ側の該システムから直接危険情報を送信するることにより伝達時間は大きく短縮される。該知的映像解析システムを監視カメラ側に設置したことにより、より早く、より広範囲に監視をする関係者に情報を送信でき安全確保をより高めた安全監視システムが可能となる。なお監視する側からは監視カメラの映像をいつでも呼び出して見ることができるのは言うまでもない。[FIG. 7] attaches a computer system in which an intelligent video analysis system is transplanted to each monitoring camera 701 at a monitoring point, and stores in the computer system actions and operations that are determined to be a preset danger as monitoring contents. The monitoring object of the image from the remote monitoring camera, for example, a person or an object represented by a car, a bicycle, etc. Video from any time before the time of the determination is sent to the monitoring room 711 and other necessary departments or people via the always-connected Internet line 702, the video, the danger information, the traced trace line, etc. To do. The video including the alarm information is transmitted to the system integration server 705 and displayed on the video display device 707, and a danger display signal such as a large red color is displayed so as to be noticeable to the monitor 710, or the danger is notified by flashing. At the same time, the video recording is recorded by the video recording device 708, and the alarm information is also sent from the automatic alarm output device 706 to a predetermined destination. The printer 709 prints related images as necessary. Information exchange via the Internet can be sent to anywhere with a personal computer such as a train cab in the vicinity of a department having an Internet receiver, without going through the system integration server. By transmitting the danger information directly from the system on the surveillance camera side, the transmission time is greatly shortened. By installing the intelligent video analysis system on the surveillance camera side, it is possible to send a safety to a related person who performs monitoring more quickly and more widely, and a safety surveillance system with higher safety can be realized. Needless to say, the surveillance camera can call and view the video of the surveillance camera at any time.

従来型の映像による安全監視システムConventional safety monitoring system with video 現行システムに知的映像解析システムの機能を追加した監視システムSurveillance system that adds intelligent video analysis system functions to the current system 踏切内監視領域の説明図Illustration of the monitoring area in a level crossing 踏切領域を出て軌道敷地内に人物が侵入A person enters the orbital premises from the level crossing area. 侵入者が軌道内に入り該侵入者が所有していた物を軌道内に放置したままで戻った場合の例An example when an intruder enters the orbit and returns with the things owned by the intruder left in the orbit 踏切が遮断している状態で人物もしくは車両等が踏切内に留まっている状態の検知とその軌跡を示す線Line indicating the detection of the state where a person or vehicle stays in the level crossing while the level crossing is blocked, and the locus 現行システムに知的映像解析システムの機能を追加した監視システムSurveillance system that adds intelligent video analysis system functions to the current system 知的映像解析システムの解析説明図Analysis diagram of intelligent video analysis system

符号の説明Explanation of symbols

101 監視カメラ 102 映像等送信装置 103 システム統合サーバ
104 記録装置 105 表示装置 106 プリンター
107 監視員 108 監視室 109 警報装置
201 監視カメラ 202 映像等送信装置 203 システム統合サーバ
204 記録装置 205 表示装置 207 車両運転士等
210 知的映像解析システム内蔵のコンピュータシステム
301 踏切バー 302 道路 303 踏切内監視領域
304 車両軌道
401 侵入者 402 踏切バー 403 侵入者の侵入軌跡表示線
404 踏切領域内 405 車両軌道(線路)
501 侵入者の軌跡表示線 502 軌道上の放置物
503 侵入者 504 車両軌道(線路) 505 踏切領域
601 踏切をくぐって侵入した人物 602 車両軌道(線路)
603 踏切バー 604 閉鎖されている踏切内を横断中の人物
605 侵入者の軌跡表示線
701 監視カメラおよび知的映像解析システム内蔵のコンピュータシステム
702 インターネット通信回線
703 インターネットサービスプロバイダへ送信
704 受信用コンピュータ
705 システム統合サーバ 706 自動警報出力装置
707 映像表示装置 708 映像記録装置 709 プリンター
710 監視員 711 監視室
801 映像の中の人物の頭部 802 手の部分 803 胴の部分
811 移動後の頭部 812 移動後の手の部分
813 移動後の胴の部分
821 映像の中の人物の存在を認識 822 移動した人物を認識
DESCRIPTION OF SYMBOLS 101 Surveillance camera 102 Image transmission device 103 System integration server 104 Recording device 105 Display device 106 Printer 107 Monitor 108 Monitoring room 109 Alarm device 201 Monitoring camera 202 Image transmission device 203 System integration server 204 Recording device 205 Display device 207 Vehicle operation 210, etc. 210 Computer system with built-in intelligent video analysis system 302 Railroad crossing 302 Road 303 Crossing monitoring area 304 Vehicle trajectory 401 Intruder 402 Railroad crossing bar 403 Intruder intrusion trajectory display line 404 Within railroad crossing area 405 Vehicle trajectory (track)
501 Intruder trajectory display line 502 Abandoned object 503 Intruder 504 Vehicle trajectory (track) 505 Railroad crossing area 601 Person who enters through the railroad crossing 602 Vehicle trajectory (track)
603 Railroad crossing bar 604 Person crossing the closed railroad crossing 605 Intruder trajectory display line 701 Computer system 702 with built-in surveillance camera and intelligent video analysis system Internet communication line 703 Send to Internet service provider 704 Receiving computer 705 System integrated server 706 Automatic alarm output device 707 Video display device 708 Video recording device 709 Printer 710 Monitor 711 Monitoring room 801 Human head in video 802 Hand portion 803 Body portion 811 Head after movement 812 After movement 813 Recognize the presence of a person in the video 822 Recognize the moved person

Claims (5)

監視カメラからの映像を送信装置によってシステム統合サーバを介して記録装置に記録し、該映像を表示装置に表示し、プリンター等で該映像をコピーして紙面等に打ち出す装置及び該映像により監視員が危険であると判断を下し警報を関連部署に伝達する警報伝達手段等を有する安全監視システムに映像の通信回線の途中に知的映像解析システムを組み込ませたコンピュータシステムを挿入することにより知的映像監視システムの機能を持つ映像監視システムを構成し、該コンピュータシステムに監視内容を記憶させ、監視地点の映像を解析して危険な行動即ち監視内容に合致した場合に、映像中の人物もしくは物体を監視対象として選び、その座標を計算して位置情報を取得し、さらに該監視対象の移動に対し追跡を続け、その追跡の軌跡を線で表し該映像に重ねて表示する機能を加えた該映像監視システムであって、該映像の視野の一部区画を指定し監視領域として設定し、監視領域に侵入もしくは逆に監視領域から外の領域に侵入した場合等も同様であるが人物もしくは物体、例えば自転車や自動車等であるが存在すれば該映像監視システムが侵入者(物)有りとし危険であると判断し、監視する該映像もしくはその危険発生を検知した時刻の任意時間の前後の映像を危険を表す表示および前記侵入者の軌跡の線図を重ねてもしくは表示部の一部分に目立つように表示し、監視員や監視の関係部署に伝達する機能を持つ安全監視システム。  A device for recording a video from a monitoring camera on a recording device via a system integration server by a transmission device, displaying the video on a display device, copying the video with a printer or the like and launching it on a paper surface, etc. It is known by inserting a computer system that incorporates an intelligent video analysis system in the middle of the video communication line into a safety monitoring system that has an alarm transmission means that determines that the video is dangerous and transmits an alarm to the relevant department. A video monitoring system having a function of a dynamic video monitoring system, storing the monitoring content in the computer system, analyzing the video of the monitoring point and matching the dangerous action, that is, the monitoring content, An object is selected as a monitoring target, its coordinates are calculated to obtain position information, and the tracking of the movement of the monitoring target is continued. Is a video surveillance system with a function of displaying the image as a line and displaying it over the video, wherein a part of the visual field of the video is designated and set as a surveillance area, and enters the surveillance area or conversely from the surveillance area The same applies to the case of entering an outside area. However, if there is a person or an object, such as a bicycle or a car, the video monitoring system determines that there is an intruder (thing) and is dangerous to monitor. The video or video before and after the time at which the occurrence of the danger is detected is displayed with the display of danger and the trajectory of the intruder superimposed or displayed prominently on a part of the display unit. Safety monitoring system with a function to communicate to related departments. 〔請求項1〕記載の知的映像監視システムに組み込ませた知的映像解析システムを使って、監視領域に侵入する人物や物体を検出し、該領域内もしくは設定を逆にして領域外にしても良いが、そこに任意設定の時間を越えて侵入し続けた場合に侵入者もしくは侵入物体であると判断し危険の警報を出し監視員や監視の関係者に伝える機能を有する安全監視システム。The intelligent video analysis system incorporated in the intelligent video surveillance system according to claim 1 is used to detect a person or an object entering the surveillance area, and set the area or the setting reverse to make it out of the area. However, a safety monitoring system that has the function of determining that it is an intruder or an intruding object and issuing a warning of danger to the monitoring personnel and monitoring personnel if it continues to invade there for an arbitrarily set time. 〔請求項1〕及び〔請求項2〕記載の知的映像監視システムに組み込ませた知的映像解析システムを使って、該映像監視システムが侵入者と判断し該侵入者が侵入を続け監視領域例えば車両軌道内等に所持品を放置もしくは放棄した場合に、この不法な所持品の放置もしくは投棄の違反行為を検出し警報を出す機能、即ち不法投棄もしくは妨害行為を検出する安全監視システム。  Using the intelligent video analysis system incorporated in the intelligent video surveillance system according to [Claim 1] and [Claim 2], the video surveillance system determines that the intruder is intruder, and the intruder continues to intrude. For example, a safety monitoring system that detects an illegal act of illegally neglecting or discarding possessed items when it is left or abandoned in a vehicle track or the like, that is, a safety monitoring system that detects illegal dumping or disturbing acts. 〔請求項3〕の機能に不法な所持品の放置もしくは投棄をした侵入者が移動した場合に映像上で追跡を続けその軌跡を線図で該映像に重ねて表示し続けて逃走方向を知らせる機能を持つことにより物品の放置による妨害行為者もしくは不法投棄者を追跡することができる安全監視システム。  [Claim 3] When an intruder who has illegally left or dumped his / her belongings illegally moves, he / she keeps tracking on the image and continuously displays the trajectory on the image to indicate the escape direction. A safety monitoring system that has the ability to track people who have been left behind or illegally dumped. 〔請求項1〕〔請求項2〕〔請求項3〕及び〔請求項4〕記載の機能を持つ監視システムであって、監視地点にある監視カメラ即ちフロントエンドに該知的映像監視システムを持つコンピュータシステムを付属せしめインターネットの機能を移植し常時接続のインターネットを使って監視員をはじめ監視の関係者のインターネットの設備を持つ全員に送るだけではなく例えば電車の運転者、運行管理者を含めた広く安全に関与する人々にインターネットを通して一度に、同時に危険等を伝達する映像情報等を送信することができる安全監視システム。  [Claim 1] [Claim 2] [Claim 3] and [Claim 4] A monitoring system having the function according to claim 1, wherein the intelligent video surveillance system is provided at a surveillance camera at a surveillance point, that is, a front end. A computer system is attached, and the functions of the Internet are ported. Using the always-connected Internet, not only is it sent to everyone who has the Internet equipment of surveillance personnel, but also includes train drivers, operation managers, etc. A safety monitoring system that can transmit video information etc. that convey dangers etc. to people who are widely involved in safety at the same time through the Internet.
JP2005261293A 2005-08-08 2005-08-08 Safety monitoring system Pending JP2007048249A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2005261293A JP2007048249A (en) 2005-08-08 2005-08-08 Safety monitoring system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2005261293A JP2007048249A (en) 2005-08-08 2005-08-08 Safety monitoring system

Publications (1)

Publication Number Publication Date
JP2007048249A true JP2007048249A (en) 2007-02-22

Family

ID=37851000

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2005261293A Pending JP2007048249A (en) 2005-08-08 2005-08-08 Safety monitoring system

Country Status (1)

Country Link
JP (1) JP2007048249A (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2008099599A1 (en) * 2007-02-14 2008-08-21 Panasonic Corporation Monitoring camera and monitoring camera control method
KR101765722B1 (en) * 2016-01-12 2017-08-08 소프트온넷(주) System and method of generating narrative report based on cognitive computing for recognizing, tracking, searching and predicting vehicles and person attribute objects and events
JP2019166996A (en) * 2018-03-23 2019-10-03 株式会社Ihi Monitoring device and monitoring method
CN113128972A (en) * 2021-04-27 2021-07-16 安徽宜人医用包装有限公司 Intelligent production management system for medical packaging bags
CN113347387A (en) * 2020-02-18 2021-09-03 株式会社日立制作所 Image monitoring system and image monitoring method
CN116886877A (en) * 2023-09-07 2023-10-13 北京博数智源人工智能科技有限公司 Park safety monitoring method and system

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10475312B2 (en) 2007-02-14 2019-11-12 Panasonic intellectual property Management co., Ltd Monitoring camera and monitoring camera control method
US9286775B2 (en) 2007-02-14 2016-03-15 Panasonic Intellectual Property Management Co., Ltd. Monitoring camera and monitoring camera control method
US9437089B2 (en) 2007-02-14 2016-09-06 Panasonic Intellectual Property Management Co., Ltd. Monitoring camera and monitoring camera control method
WO2008099599A1 (en) * 2007-02-14 2008-08-21 Panasonic Corporation Monitoring camera and monitoring camera control method
US9870685B2 (en) 2007-02-14 2018-01-16 Panasonic Intellectual Property Management Co., Ltd. Monitoring camera and monitoring camera control method
US10861304B2 (en) 2007-02-14 2020-12-08 Panasonic I-Pro Sensing Solutions Co., Ltd. Monitoring camera and monitoring camera control method
KR101765722B1 (en) * 2016-01-12 2017-08-08 소프트온넷(주) System and method of generating narrative report based on cognitive computing for recognizing, tracking, searching and predicting vehicles and person attribute objects and events
JP2019166996A (en) * 2018-03-23 2019-10-03 株式会社Ihi Monitoring device and monitoring method
JP7212455B2 (en) 2018-03-23 2023-01-25 株式会社Ihi Monitoring device and monitoring method
CN113347387A (en) * 2020-02-18 2021-09-03 株式会社日立制作所 Image monitoring system and image monitoring method
CN113128972A (en) * 2021-04-27 2021-07-16 安徽宜人医用包装有限公司 Intelligent production management system for medical packaging bags
CN113128972B (en) * 2021-04-27 2023-10-10 安徽宜人医用包装有限公司 Intelligent production management system for medical packaging bags
CN116886877A (en) * 2023-09-07 2023-10-13 北京博数智源人工智能科技有限公司 Park safety monitoring method and system
CN116886877B (en) * 2023-09-07 2023-11-14 北京博数智源人工智能科技有限公司 Park safety monitoring method and system

Similar Documents

Publication Publication Date Title
KR101766305B1 (en) Apparatus for detecting intrusion
JP2007195130A (en) Video monitoring system
US8346056B2 (en) Graphical bookmarking of video data with user inputs in video surveillance
EP1775692A2 (en) Safety procedure and device for vehicle junctions
JP2006287884A (en) Intelligent video monitoring system
KR101229945B1 (en) System for alarm and detection accident of tunnel
KR101036947B1 (en) The automatic guard system to prevent the crime and accident using computer video image analysis technology
JP2007048249A (en) Safety monitoring system
JP2012014302A (en) Monitoring device
KR101936837B1 (en) Map-based intelligent video integrated surveillance system and monitoring method using the same
MX2008007319A (en) Remote area monitoring system.
KR20080044936A (en) Watching control system having an auto picture pop-up function and controlling method for the same
KR20140134457A (en) Safety Accident Management System Of Bridge
JP2007089105A (en) Walk monitoring system
KR100690279B1 (en) Multipurpose video image detection system
JP2006232024A (en) Safety monitoring system, safety device, and alarm device
JP2003063402A (en) Safety monitor device for platform in station
JPH0248263A (en) Railroad-crossing monitor
KR100661264B1 (en) Automatical danger detecting system at railroad crossing using thermal image camera
JPH10269492A (en) Vehicle monitoring device
JP2003272086A (en) Tunnel inside monitoring system, program and recording medium
JP7367781B2 (en) Monitoring system, monitoring device, monitoring method, and program
KR100497311B1 (en) Method for automatically detecting and alarming unexpected accident while boarding or alighting from subway and train at platform
KR100699596B1 (en) Image processing alarm system for automatically sensing unexpected accident at railroad crossing
JP2006287883A (en) Video monitoring system