JP2011215767A - Server device, method of using security camera images, program for using security camera images, and security camera system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To effectively use camera images from security cameras installed in various places, instead of using manpower, so as to contribute to crime or accident prevention.SOLUTION: One or more pieces of comparison video information for detecting the occurrence of one or more predetermined events, which are prepared in advance in a registered image DB 132 and a crime and accident pattern image DB 133, and video information received by a communication unit 101 from one or more security camera devices are matched by a registered image matching unit 103 and a crime and other pattern matching unit 104; and when the occurrence of the predetermined event near an installation place of the security camera device is detected, a notification processor notifies a predetermined notification destination of the occurrence of the event.

Description

  The present invention relates to an apparatus, a method, a program, and a system for effectively using video information from one or more security camera devices installed in, for example, a shopping street, a store entrance / exit, and a home entrance.

  In stores where an unspecified number of people come and go, a security camera (surveillance camera) is provided mainly for the purpose of crime prevention, and the occurrence of abnormalities can be detected quickly by monitoring the video taken by the security camera. In some cases, the intruders are identified and cleared. In recent years, security cameras have also been installed in general households to take pictures of the area near the doorway, to prevent the intrusion of suspicious persons, and to help identify and detect suspicious persons. It has come to be. In addition, installation of a plurality of surveillance cameras in a relatively large area such as a downtown area or a shopping street to help prevent crimes in the area has also been performed in various places.

  Security cameras can be reduced in size, weight, and cost, and can be introduced relatively easily and easily. Improvements have been made. For example, Patent Document 1 discloses an invention relating to sharpening of an image taken by a security camera. Patent Document 2 discloses an invention relating to long-time recording of images taken by a security camera. Various other improvements have also been made, and it has become possible to take a clearer picture with a security camera and record it on a recording medium for a long time.

JP 2010-056724 A JP 2009-296266 A

  By the way, the conventional usage mode of a security camera is roughly divided into two usage modes. One is to constantly monitor images from security cameras installed at a plurality of places, for example, from a security room, and to allow a monitor to detect the occurrence of an abnormality at the plurality of places in real time. The other is to record a video from a security camera for a long time, and when an intruder or the like is generated, the recorded video is used for identification and clearance of the intruder or the like.

  In the former case, if an abnormality is to be detected quickly, the observer must always monitor the video from the security camera. For this reason, it can be realized in the case of an apartment house or a company where security guards always live, but it is generally impossible to arrange a supervisor at a small store or a private house. In the latter case, the occurrence of abnormality cannot be detected in real time. Therefore, the advantage of the latter usage mode is that an image useful for identifying or checking an intruder or the like after an abnormality has occurred can be obtained. For this reason, it can be said that the video from the security camera, which has been installed in various places and has been able to record a clear video for a longer time, has not been fully utilized.

  In view of the above, the present invention enables effective use of video from security cameras installed in various locations without bothering humans and contributes to prevention of incidents and accidents in a wider area. With the goal.

In order to solve the above problem, a server device according to claim 1 is provided:
Storage means for storing one or more reference video information for detecting the occurrence of one or more predetermined events;
Receiving means for receiving video information obtained by each of one or more security camera devices installed in the city taking a video of the surroundings and transmitted through a predetermined communication network;
The video information received through the receiving means is compared with each of the one or more reference video information stored in the storage means to detect the occurrence of the predetermined event near the installation position of the security camera device Detecting means for
And executing means for executing processing determined according to the detected event when the detection means detects the occurrence of any one or more of the predetermined events.

  According to the server device of the first aspect of the present invention, each of the video information from the one or more security camera devices received by the receiving unit and the occurrence of one or more predetermined events prepared in advance in the storage unit are generated. Each of the one or more pieces of reference video information for detection is matched (verified) by the detection means, and the occurrence of a predetermined event near the installation position of the security camera device is detected. When the detection unit detects the occurrence of a predetermined event in the vicinity of the installation position of the security camera device, the execution unit executes a process determined according to the detected event.

  As a result, a predetermined event in the vicinity of the security camera device is automatically detected. When the occurrence of the predetermined event is detected, a process corresponding to the detected event, for example, a predetermined process such as notifying the notification destination corresponding to the detected event of the detection of the event is executed. The Therefore, it is possible to effectively utilize the images from the security cameras installed in various places without intervention, and contribute to the prevention of incidents and accidents in a relatively large area.

  According to the present invention, it is possible to effectively utilize images from security cameras installed in various places without involving human hands, and to contribute to prevention of incidents and accidents in a relatively large area. It also contributes to the resolution of incidents and accidents.

It is a figure for demonstrating one Embodiment of the security camera system by this invention. It is a figure for demonstrating an example of the corresponding area of the security camera system of embodiment. It is a block diagram for demonstrating the structural example of each security camera system 2, 3, and 4. FIG. It is a block diagram for demonstrating the structural example of each security camera system 2, 3, and 4. FIG. 3 is a block diagram for explaining a configuration example of a video management server 1. FIG. It is a figure for demonstrating the example of the data managed in camera installation position DB131. It is a figure for demonstrating the example of the data managed in registration image DB132. It is a figure for demonstrating the example of the data managed in case / accident pattern image DB133. It is a figure for demonstrating the example of the data managed in notification destination DB134. 6 is a diagram for explaining an example of data managed in a tracking data file 135. FIG. 5 is a flowchart for explaining processing performed in the video management server 1; It is a flowchart following FIG.

  Hereinafter, an embodiment of the apparatus, method, program, and system of the present invention will be described with reference to the drawings.

[Outline of security camera system]
FIG. 1 is a diagram for explaining an embodiment of a security camera system according to the present invention. The security camera system of this embodiment includes a security camera video management server (hereinafter abbreviated as video management server) 1, a store security camera system 2 (1), 2 (2),. The systems 3 (1), 3 (2),... And the street security camera systems 4 (1), 4 (2),.

  The store security camera systems 2 (1), 2 (2), ... are security cameras installed near the entrances and exits of various stores (for example, convenience stores, gas stations, banks, etc.) where an unspecified number of people enter and exit. It consists of the attached device. In this embodiment, each of the store security camera systems 2 (1), 2 (2),... Has a shooting range not only near the entrance of the installed store but also a part of the road around it, for example, It can also take pictures of road traffic.

  Further, the home security camera systems 3 (1), 3 (2),... Are composed of a security camera installed near an entrance / exit such as the entrance of each home and its associated devices. In this embodiment, each of the home security camera systems 3 (1), 3 (2),... Also covers not only the vicinity of the entrance of the installed house but also a part of the road around it, for example. It is also possible to photograph the traffic condition of the road.

  The security camera systems 4 (1), 4 (2), etc. in the city are, for example, security cameras and their associated devices provided for crime prevention by, for example, residents' associations, shopping associations, etc. It consists of Each of the street security camera systems 4 (1), 4 (2),... Is highly public, and can widely shoot the state of the installed shopping street and downtown.

  Normally, each of the store security camera systems 2 (1), 2 (2),... Is provided as a closed system in the installation store. Moreover, each of the home security camera systems 3 (1), 3 (2),... Is provided as a closed system in the installed home. However, as described above, each of the store security camera systems 2 (1), 2 (2),... Also covers a part of the surrounding road including the vicinity of the entrance of the installed store. Similarly, each of the home security camera systems 3 (1), 3 (2),... Covers not only the vicinity of the entrance of the installed house but also a part of the surrounding road.

  Therefore, in addition to the images shot by the street security camera systems 4 (1), 4 (2),..., The store security camera systems 2 (1), 2 (2),. If the video taken in 1), 3 (2),... Can also be used, a security camera system for a wide area can be constructed.

  FIG. 2 is a diagram for explaining an example of a corresponding area of the security camera system according to this embodiment. In FIG. 2, the black circle mark indicates the camera installation position, and the arrow from the black circle mark indicates the shooting direction. In FIG. 2, in the area above the station, the security camera system 2 (1), 2 (2), 2 (3), 2 (4), 2 (5) 2 (6) and 2 (7) are provided. Home security camera systems 3 (1), 3 (2), 3 (3), and 3 (4) are provided at positions slightly further from the station than the store security camera system described above. Furthermore, the street security camera system 4 (1), 4 (2), 4 (3), 4 (4), 4 (5) is provided on the main street of the shopping street extending north from the station. .

  Further, in FIG. 2, in the area below the station, the security camera systems 2 (8), 2 (9), 2 (10), 2 (11), 2 (12 ), 2 (13). Home security camera systems 3 (5), 3 (6), 3 (7), 3 (8), and 3 (9) are provided at positions slightly further away from the station than the store security camera system described above. ing. Furthermore, on the main street of the shopping street that extends south from the station, there are city security camera systems 4 (5), 4 (6), 4 (7), 4 (8), and 4 (9). .

  Therefore, the security camera system of this embodiment is a security camera system provided at various positions as shown in FIG. 2, and is mainly used for monitoring only in the vicinity of the installation location. 2 (1), 2 (2), ..., home security camera systems 3 (1), 3 (2), ..., and street security camera systems 4 (1), 4 () that cover a relatively wide area. 2), ... are linked. In this way, as shown in FIG. 2, for example, a wide-area security camera system in which a relatively wide area is a monitoring target area is constructed. In this way, various security camera systems and the video management server 1 are connected to form a security camera system having a relatively large area as a monitoring area.

  In the video management server 1, video information transmitted from each security camera system is received and the occurrence of a predetermined event is detected by using this video information, which is useful for prevention of incidents and accidents. Furthermore, when an incident or accident occurs, the video information from each security camera system can be used to track an actor (such as an incident or an offender of the accident).

  Specifically, the video management server 1, as will be described in detail later, is image data of the face of an elderly person with dementia, image data of a face of an infant or child, face image data of a patient with sleepwalking, and a criminal (designated criminal) Registration and reception of the image data of the face as reference video information. In addition, image data such as the face portions of pets such as dogs and cats that have escaped are accepted and managed as reference video information. When the video management server 1 finds a person (detection target person) of the registered image data or an animal (detection target animal) of the registered image data in the video information from the security camera system, the communication network 100 To a predetermined notification destination (notification destination).

  As shown in FIG. 1, the notification destination in this case is, for example, a telephone of a contractor 6 that registers image data of a security company 5 or a demented elderly person who has a contractual relationship with the operating organization of the video management server 1. For example, a personal computer. It is also possible to notify the contractor's mobile phone terminal 10 (1) through the base station 9 (1).

  In addition, a person registered as a collaborator can be requested to protect the detected person to be detected through his / her mobile phone terminal. In addition, by knowing the location of the cooperator's home from the latitude, longitude information and address information, the cooperator living in the vicinity of the security camera including the image of the detection target person and the animal to be detected is detected. You can also ask for animal protection.

  In addition, for example, when a wanted person is detected, a message for calling attention can be transmitted to a person registered as an alarm destination. In addition, by ascertaining the home location of the alarm destination from latitude, longitude information and address information, a warning message is sent to the alarm destination living in the vicinity of the security camera including the image of the wanted crime. be able to.

  The current position of the collaborator and the alarm destination is detected by using a GPS (Global Positioning System) function that has recently been installed in a mobile phone terminal, and the video management server 1 periodically detects this. By receiving and managing, the current position of cooperators and alarm destinations is grasped. And it is also possible to specify a collaborator and a warning destination based on this current position.

  The video management server 1 also notifies the digital signage center 7 when a detection target person or a detection target animal is detected by a security camera. In this case, the digital signage center 7 displays a message for requesting protection or a message for calling attention on a digital signage device such as a liquid crystal display device provided in the city, and cooperates with general users. You can also request or call attention.

  The video management server 1 notifies a CATV (Community Antenna TeleVision) company 8 when a detection target person or a detection target animal is detected by a security camera. In this case, the CATV company transmits a broadcast signal to be transmitted including a message requesting protection or a message for calling attention, and displays the television receiver of a home or the like that is contracted with the CATV company. By displaying the message on the screen, it is possible to request cooperation from the user or call attention.

  Further, the video management server 1 accepts and manages registration of the case image, such as a failure case, a snatching case, a traffic accident, and the pattern image of the accident as reference video information. When the image management server 1 detects the occurrence of an incident or accident due to the presence of an image pattern corresponding to the registered pattern image in the video information from the security camera system, the video management server 1 performs a predetermined notification through the communication network 100. Notify (inform) first.

  The notification destination in this case is the same as when detecting the detection target person or the detection target animal described above, the security company 5, the contractor 6 (including the cooperator and the alarm destination), the mobile phone terminal 10 (such as the contractor) ( 1) Signage center 7, CATV company 8, etc. In addition, a police station and a fire station (not shown) are also notified. In addition, as described above, it accepts registration of the current location of the alarm destination detected by the GPS function of the alarm destination mobile phone terminal, and gives priority to the alarm destination near the site where the incident or accident occurred You can also do things.

  In FIG. 1, only one security company 5, contractor 6, signage center 7, CATV company 8, base station 9 (1), and mobile phone terminal 10 (1) are shown. Of course, there may be more than one. In particular, there are as many contractors 9 and mobile phone terminals 10 (1) as there are contractors and registrants, and there are many base stations on the mobile phone network. Further, as described with reference to FIG. 2, a large number of security camera systems for shops, home security camera systems, and street security camera systems are provided and are connected through the communication network 100.

  In this embodiment, the communication network 100 is mainly assumed to be the Internet. However, as a connection path to the Internet, a PSTN, ISDN, a mobile phone network including a base station, and the like are included in part. Furthermore, the communication network 100 may include a wireless local area network (LAN) using, for example, [Wifi] or “Wimax (Worldwide Interoperability for Microwave Access)” that has been used in recent years. PSTN stands for public switched telephone network and is an abbreviation for Public Switched Telephone Networks. ISDN stands for Integrated Services Digital Network and is an abbreviation for Integrated Services Digital Network.

[Configuration example of security camera system]
Next, the security camera system for stores 2 (1), 2 (2),..., The home security camera systems 3 (1), 3 (2),. ), 4 (2),... Each of these security camera systems has the same configuration. Therefore, here, the store security camera systems 2 (1), 2 (2),... Are collectively referred to as the store security camera system 2, and the home security camera systems 3 (1), 3 (2),. Are collectively referred to as a security camera system 3 for a city, and the security camera systems 4 (1), 4 (2),... For a city are collectively referred to as a security camera system 4 for a city, and a basic configuration example thereof will be described.

  FIGS. 3 and 4 are block diagrams for explaining a configuration example of each security camera system 2, 3, 4. FIG. 3 shows an example in which a security camera system is configured by wired connection. The security camera 201, the digital recorder 202, the television monitor 203, and the transmitter 204 are considerably connected between the security camera 201 and the digital recorder 202. Connected by wire. In the security camera system shown in FIG. 3, the security camera 201 is generally installed outdoors, and the digital recorder 202, the television monitor 203, and the transmitter 204 are generally installed indoors.

  The security camera 201 is for outdoor installation, and has been developed as a video camera dedicated to a security camera system capable of continuously shooting a relatively wide range of images. The basic configuration of the security camera 201 is the same as that of a general video camera, such as a lens unit, an image sensor such as a CCD (Charge Coupled Device) image sensor or a CMOS (Complementary Metal Oxide Semiconductor) image sensor, An A / D (Analog / Digital) converter for converting the obtained analog video signal into a digital signal is provided. The security camera 201 can convert a video signal (video information) obtained by shooting a video in a target range into a digital video signal, and supply the digital video signal to a subsequent digital recorder 202 via a cable. It is.

  The digital recorder 202 uses a hard disk having a storage capacity of at least several hundred GB (gigabytes) to several TB (terabytes) as a recording medium. The digital recorder 202 receives the digital video signal from the security camera 201, compresses the data and writes (records) it on the hard disk, and the digital video signal from the security camera 201 to the television monitor 203 at the subsequent stage. An output unit for outputting.

  The digital recorder 202 also includes a reading unit that reads a digital video signal recorded on a hard disk, a decompressing unit that decompresses the read digital video signal and restores the digital video signal before compression, and a restored digital video signal ( A reproduction video signal output unit for outputting the reproduction video signal) to a television monitor or the like.

  Furthermore, the digital recorder 202 also has an output unit that outputs, for example, a digital video signal from the security camera 201 compressed in data to the transmitter 204. Here, the digital video signal that has been subjected to data compression is output to the transmitter 204. However, it is also possible to output the digital video signal to the transmitter 204 without data compression.

  The television monitor 203 is an existing television receiver, for example. The television monitor 203 includes at least an external input terminal that receives an input of a digital video signal, a video processing unit that forms a video signal for display from the digital video signal from the external input terminal, and a video signal from the video processing unit And a display element such as an LCD (Liquid Crystal Display) that displays an image corresponding to the display on the display screen.

  The transmitter 204 is supplied with a digital video signal of a video shot by the security camera 201 supplied from the digital recorder 201. Then, the transmitter 204 forms a digital video signal for transmission in a format to be transmitted from the supplied digital video signal to the video management server 1 via the communication network 100, and transmits the digital video signal to the video management server 1 via the communication network 100. It has the function to transmit to. That is, the transmitter 204 realizes a function of transmitting a video signal of a video taken by the security camera 201 to the video management server 1 through the communication network 100.

  Thus, in the case of the security camera system shown in FIG. 3, the video signal (video information) obtained by taking the video with the security camera 201 is recorded on the hard disk of the digital recorder 202, and the video formed by the video signal. Can be monitored by the television monitor 203. Further, in the case of the security camera system shown in FIG. 3, a video signal (video information) taken by the security camera 201 is sent to the communication network 100 through the digital recorder 202 and the transmitter 204 in real time, and the video management server. 1 can also be transmitted.

  FIG. 4 shows an example in which a security camera system is configured by wireless connection. In the security camera system shown in FIG. 4, parts that are configured in the same manner as the security camera system shown in FIG. 3 are given the same reference numerals, and descriptions thereof are omitted because they are duplicated. As shown in FIG. 4, a transmitter 201 </ b> S is connected to the rear stage of the security camera 201, and a receiver 202 </ b> R is connected to the front stage of the digital recorder 202. In the security camera system shown in FIG. 4, the security camera 201 and the transmitter 201S are installed outdoors, and the receiver 202R, the digital recorder 202, the television monitor 203, and the transmitter 204 are installed indoors. It is common.

  In FIG. 4, a transmitter 201S modulates a digital video signal from the security camera 201 to form a signal for wireless transmission, and wirelessly transmits the signal to a receiver 202R. The receiver 202R receives the radio transmission signal transmitted from the radio 201S, demodulates the signal, extracts the transmitted digital video signal, and supplies the digital video signal to the digital recorder 202.

  Accordingly, in the case of the security camera system shown in FIG. 4, the video signal (video information) obtained by the security camera 201 is supplied to the digital recorder 202 by wireless communication, and this is stored in the hard disk of the digital recorder 202. While recording, the video formed by the video signal can be monitored by the television monitor 203. Further, in the case of the security camera system shown in FIG. 4, a video signal (video information) taken by the security camera 201 is wirelessly transmitted to the digital recorder 202 in real time, and the communication network is transmitted through the digital recorder 202 and the transmitter 204. It can also be sent to 100 and sent to the video management server 1.

  The security camera system having the configuration shown in FIG. 3 is often used mainly in stores and homes, and the security camera system having the configuration shown in FIG. 4 is mainly used in street security camera systems. However, the difference is whether the security camera 201 and the digital recorder 202 are connected by wire or wirelessly. Therefore, the security camera system shown in FIGS. 3 and 4 can be applied to various security camera systems.

[Configuration Example of Video Management Server 1]
Next, a configuration example of the video management server 1 shown in FIG. 1 will be described. FIG. 5 is a block diagram for explaining a configuration example of the video management server 1 according to this embodiment. As shown in FIG. 5, the video management server 1 includes a communication unit 101, a security camera video storage unit 102, a registered image matching unit 103, an incident pattern matching unit 104, a notification processing unit 105, and a tracking processing unit. 106.

  In addition, the video management server 1 includes an input operation unit 121, an input / output terminal 122, and an external interface (hereinafter abbreviated as an external I / F) 123 as a direct information receiving end. Further, the video management server 1 includes a camera installation position DB 131, a registered image DB 132, an incident / accident pattern image DB 133, a notification destination DB 134, and a tracking data file 135 as various data storage units. Here, the term “DB” is used as an abbreviation for database. The video management server 1 also includes a control unit 110.

  In the video management server 1, the communication unit 101, the security camera video storage unit 102, the registered image matching unit 103, the incident pattern matching unit 104, the notification processing unit 105, and the tracking processing unit 106 are for realizing important functions. The main part of In order to simplify the description of these main parts, first, the peripheral parts of the main parts will be described.

  As shown in FIG. 5, the control unit 110 includes a CPU (Central Processing Unit) 111, a ROM (Read Only Memory) 112, a RAM (Random Access Memory) 113, an EEPROM (Electrically Erasable and Programmable ROM) 114, and a CPU bus 115. It is the microcomputer comprised connected through. The control unit 110 can control each unit of the video management server 1.

  Here, the CPU 111 is a main body of control, reads out and executes a program stored in the ROM 112 or the EEPROM 114, forms a control signal for each part, and supplies it to each related part. Receive data from each unit and process it. As described above, the ROM 112 stores various processing programs executed by the CPU 111 and data necessary for processing in advance.

  The RAM 113 is mainly used as a work area in various processes, such as temporarily storing intermediate results of the processes. The EEPROM 114 is a rewritable so-called non-volatile memory. In addition to data to be retained even when the power is turned off, for example, a new program for adding a function, various setting parameters, etc. Various information such as an IP address assigned to is stored and held.

  As described above, the control unit 110 has a microcomputer configuration. Therefore, the control unit 110 executes the functions of the registered image matching unit 103, the case pattern matching unit 104, the notification rib 105, and the tracking processing unit 106 indicated by double lines in FIG. It can also be realized as a function of the control unit 110 by a program.

  The operation input unit 121 enables manual input of information by the operator. Specifically, the operation input unit 107 includes one or both of a keyboard having various operation keys such as alphabet keys and a pointing device such as a so-called mouse. Of course, various pointing devices such as a touch panel, an input device composed of a tablet and a touch pen can be used.

  The input / output terminal 122 is a part that accepts connection of an external device and enables data input from the external device and data output from the own device to the external device. The external I / F 123 converts data from an external device received through the input / output terminal 122 into data in a format that can be processed by the own device, and supplies the data to the control unit 110. Also, the external I / F 123 converts the data from the control unit 110 into data in a format to be output to the external device, and outputs this to the external device connected to the input / output terminal. The input / output terminal 122 and the external I / F 123 are parts realized as a digital interface that conforms to a predetermined standard such as USB (Universal Serial Bus).

  Each of the camera installation position DB 131, the registered image DB 132, the incident / accident pattern image DB 133, the notification destination DB 134, and the tracking data file 135 is formed, for example, on a hard disk having a large storage capacity. Therefore, each of the camera installation position DB 131, the registered image DB 132, the incident / accident pattern image DB 133, the notification destination DB 134, and the tracking data file 135 is realized as a hard disk drive configuration.

  The camera installation position DB 131, the registered image DB 132, the incident / accident pattern image DB 133, the notification destination DB 134, and the tracking data file 135 may be formed on different hard disks as shown in FIG. Further, the camera installation position DB 131, the registered image DB 132, the incident / accident pattern image DB 133, the notification destination DB 134, and the tracking data file 135 may be formed on the same hard disk. Further, the notification destination DB 134 and the tracking data file having a relatively small amount of data to be stored can be created on a recording medium having a relatively small storage capacity such as a semiconductor memory. Hereinafter, specific examples of data managed in each of the camera installation position DB 131, the registered image DB 132, the incident / accident pattern image DB 133, the notification destination DB 134, and the tracking data file 135 will be described.

[Example of data in the camera installation position DB 131]
First, an example of data managed by the camera installation position DB 131 will be described. The camera installation position DB 131 manages information related to the installation position of the security camera of the security camera system that transmits the video information to the video management server 1. FIG. 6 is a diagram for explaining an example of data managed in the camera installation position DB 131 of this embodiment.

  In this embodiment, the camera installation position DB 131 manages the identification ID, type, installation position, shooting direction, and neighboring camera information for each security camera of the security camera system that transmits video information to the video management server 1. . The “identification ID” is information that can uniquely identify each security camera. “Type” is information indicating the installation subject of the security camera. Specifically, the “type” is a security camera for a street security camera system installed by a neighborhood association or a shop association, a security camera for a security camera system for a store installed in a store, or a household part installed in a private house. This information identifies whether the camera system is a security camera.

  “Installation position” is information indicating the installation position of the security camera, and includes, for example, longitude information and latitude information. The longitude information and latitude information in this case can be acquired by designating the installation position of the security camera on the map at a site that provides map data, for example. In addition, the “installation position” can be obtained, for example, by measuring at the installation location using a portable GPS device. In addition, although it is possible to use an address instead of longitude information and latitude information, longitude information and latitude information are more accurate. The “shooting direction” is information indicating the shooting direction of the security camera. The “neighboring camera” information includes information for identifying other security cameras in the vicinity of the security camera specified by the identification ID.

  In the example of the camera installation position DB data shown in FIG. 6, the reference symbol attached to the security camera system in FIG. 2 is used as the identification ID for the sake of simplicity. As shown in FIG. 6, the security camera with the identification ID “4 (1)” is one of the street security camera systems 4 installed by the residents' association as described with reference to FIG. is there. As shown in FIG. 2, the security camera with the identification ID “4 (1)” is installed in front of the station on the north side of the station XX, and the shooting direction is the northwest direction. Furthermore, in the vicinity of the security camera with the identification ID “4 (1)”, as shown in FIG. 2, the security camera 2 (3) is located on the east side, and the security camera 2 (2) is located on the west side. It is shown that a security camera 4 (5) is installed on the south side with a station in between, and a security camera 4 (2) is installed on the north side.

  As shown in FIG. 6, the security camera with the identification ID “3 (1)” is one of the home security camera systems 3 installed in the home as described with reference to FIG. As shown in FIG. 2, the security camera with the identification ID “3 (1)” is installed in the west part on the north side of the station XX, and the shooting direction is shown to be northeast. . Further, in the vicinity of the security camera with the identification ID “3 (1)”, as shown in FIG. 2, the security camera 2 (5) is installed on the east side, and the security camera installed on the west side in FIG. There is no security camera 2 (1) on the south side and a security camera 4 (4) on the north side.

  Also, as shown in FIG. 6, the security camera with the identification ID “2 (1)” is one of the security camera systems 2 for stores installed in the store, as described with reference to FIG. As shown in FIG. 2, the security camera with the identification ID “2 (1)” is installed in the west part on the north side of the station XX, indicating that the shooting direction is southeast. . Further, in the vicinity of the security camera with the identification ID “2 (1)”, as shown in FIG. 2, the security camera 2 (2) is installed on the east side and the security camera installed on the west side in FIG. It is shown that a security camera 2 (8) is installed on the south side with a track on it, and a security camera 3 (1) is installed on the north side.

[Example of registered image DB 132 data]
Next, an example of data managed by the registered image DB 132 will be described. The registered image DB 132 receives registration of image data used as reference video information such as image data of the face of the person to be monitored (detection target person) and image data of the face part of the animal to be monitored (detection target animal). And memorize it. Here, the person to be monitored is, for example, a person who is dangerous if one person such as an elderly person with dementia, an infant or a child goes out, or a person who wants to arrange a person. And the image data of the face of demented elderly people, infants and children to be monitored is provided by the family of these persons. For example, the image data of the face of the wanted criminal is provided by the police.

  The monitored animals are, for example, so-called pets raised by users such as dogs that have escaped and cats that have stopped returning. For example, the image data of the face portion of a pet such as a dog that has escaped or a cat that has not returned is provided by the owner of the pet.

  It should be noted that the image data of the face of the person to be searched and the image data of the face of the animal to be searched for, for example, those transmitted from the other party through the communication network 100 are received through the communication unit 101 to be described later. It can be registered in the registered image DB 132 through the control unit 110. Further, the image data of the face of the search target person and the image data of the face of the search target animal, for example, are received through the input / output terminal 122 and the external I / F 132 and registered in the registered image DB 132 through the control unit 110. You can also

  FIG. 7 is a diagram for explaining an example of data managed in the registered image DB 132 of this embodiment. In this embodiment, in the registered image DB 132, the registration ID, type, notification destination, and individual information are added to the image data of the search target person or animal face provided as described above and managed. Is done. In the case of the example shown in FIG. 7, registration is performed with the supply of the image data 1 for the front face and the image data 2 for the face side for the purpose of speeding up the image matching process and improving the accuracy described later. Like to do.

  However, recent image analysis techniques have improved. For example, when face-side image data can be generated from face-front image data, face-side image data is not essential. On the contrary, if the image data of the face side can be generated if the image data of the face side is present, it is sufficient that either the image data of the face front or the image data of the face side exists.

  Then, the “registration ID” added to the image data such as the search target person or the face of the animal can uniquely identify the image data. The “type” is information indicating what the search target is, such as an elderly person, a wanted criminal, an infant, a child, a pet (dog), and the like. Further, “notification destination” is information for specifying a destination to be notified of when a search target person or animal is caught by a security camera. Specifically, as shown in FIG. 1, the “notification destination” is a registration ID of a notification destination that specifies a notification destination such as a contractor, a predetermined security company, or the police. The “individual information” is information such as the name, age, height, weight, characteristics, and other information of the person to be searched, and the pet name, type, length, weight, characteristics, etc. of the animal to be searched.

  In the case of the example shown in FIG. 7, as the registered image with the registration ID “0001”, the person to be searched is an elderly person, the notification destination is the contractor and the security company, and the search is performed as individual information. Image data (0001-1) of the front face to which the name, age, height, weight, and features of the target person are added and image data (0001-2) of the face side are registered. In addition, as shown in FIG. 7, in the registered image DB 132, image data about a person or animal to be searched for such as a wanted criminal, a child, or a pet (dog) is added with a registration ID, a notification destination, and individual information. Has been registered.

[Example of incident / accident pattern image DB data]
Next, an example of data managed by the incident / accident image pattern DB 133 will be described. The incident / accident pattern image DB 133 stores and holds pattern image data used as reference video information, such as an incident to be detected, an incident for detecting the occurrence of an accident, or one or more pattern image data of an accident. .

  Here, the cases and accidents that are the detection target actions are obstacle cases, snatching cases, traffic accidents, and the like. And, for example, the pattern image in the case of a disability case such as hitting a person is a person who has been beaten or fallen down (a victim) ) And exist. Also, the pattern image in the case of an injury case in which a person cuts a person with a blade includes a person (perpetrator) who swings his arm with the blade and a person (victim) who escapes.

  Moreover, the pattern image in the case of a traffic accident includes, for example, a state in which one or more automobiles are deformed and stopped, or there are persons or objects (bicycles or motorcycles) that have fallen near the automobile. To do. As described above, the previously created pattern images of the cases and accidents to be detected are stored and held in the case / accident pattern image DB 133. For example, the case and accident pattern images created by the police and related organizations are used.

  The incident or accident pattern image, which is the action to be detected, is transmitted through the communication network 100 from the other party such as the police through the communication unit 101 to be described later. It can be registered in the pattern image DB 133. It is also possible to accept an incident that is an action to be detected or a pattern image of itself through the input / output terminal 122 and the external I / F 132 and register it in the incident / accident pattern image DB 133 through the control unit 110.

  FIG. 8 is a diagram for explaining an example of data managed in the incident / accident pattern image DB 133 of this embodiment. In this embodiment, the incident / accident pattern image DB 133 is managed by adding a registration ID, type, notification destination, and remarks information to the pattern image data of the detection target action provided as described above. .

  The “registration ID” added to the pattern image data of the detection target action can uniquely identify the pattern image data. In the case of this example, the registration ID of the pattern image data is the head such as “J001” in order to distinguish it from image data of a person to be monitored such as an elderly person managed by the registration image DB described above. The letter “J” of the alphabet is added.

  The “type” is information indicating what kind of detection target actions are, for example, traffic accident 1, traffic accident 2, failure 1, and failure 2. Further, “notification destination” is information for specifying a destination to notify when the occurrence of the detection target action is caught by the security camera. Specifically, the “notification destination” is a registration ID of a notification destination that mainly identifies a notification destination such as a police or a fire brigade.

  The remark information is information indicating the details of the action to be detected in more detail. For example, a pattern image with a “type” of traffic accident 1 indicates that it is a pattern image of a car-to-car traffic accident, or a pattern image with a “type” of failure 1 indicates that it is a failure case due to a beating. And so on. In addition, various information can be added as remark information.

  In the case of the example shown in FIG. 8, as the pattern image with the registration ID “J001”, “Type” is the traffic accident 1, “Notification destination” is the police, and “Remarks information” is the vehicle A pattern image (001) to which information indicating that it is a pattern image of a car traffic accident is added is registered. In addition, as shown in FIG. 8, pattern images such as traffic accident 2, traffic accident 3,..., Failure 1,.

[Example of data in notification destination DB 134]
Next, an example of data managed by the notification destination DB 134 will be described. The notification destination DB 134 is a notification destination when a monitoring target person (detection target person) or a monitoring target animal (detection target animal) is detected by a security camera, or when an occurrence of a detection target action is detected by a security camera. It manages information about. FIG. 9 is a diagram for explaining an example of data managed in the notification destination DB 134 of this embodiment.

  In this embodiment, the notification destination DB 134 notifies that a detection target person (detection target person) or a monitoring target animal (detection target animal) is detected or when a detection target action is detected. Registration ID, type, name, telephone number, and e-mail address (e-mail address) are managed for each destination (notification destination).

  “Registration ID” is information that can uniquely identify a registered notification destination. “Type” is information indicating whether the notification destination is the police, the fire department, the contractor, the alarm destination, or the collaborator. “Name” is information indicating the name (name) of the notification destination. “Telephone number” indicates the telephone number of the notification destination, and “E-mail” indicates the e-mail address of the notification destination.

  In the example shown in FIG. 9, the telephone number and e-mail address of the police station are registered as the notification destination having the registration ID “001”, and the notification destination having the registration ID “002” is ○ The telephone number and e-mail address of the fire department are registered. Further, the telephone number and e-mail address of XX security company are registered as the notification destination with the registration ID “003”, and the telephone number and e-mail address of the security company are registered as the notification destination with the registration ID “004”. It is registered.

  Further, in the case of the example shown in FIG. 9, the names of a plurality of contractors that are configured to register image data of persons to be monitored such as elderly people with dementia, infants, and children, and animals to be monitored, in the registered image DB 132. Phone number and e-mail address are registered. In addition, as will be described in detail later, the names of one or more alarm destinations, telephone numbers, and e-mail addresses that receive notification of incidents and accidents, for example, monitored persons, monitored animals, etc. At the time of discovery, the names, telephone numbers, and e-mail addresses of one or more cooperators who are requested to protect the detection target are also registered.

  In addition to this, although not shown in FIG. 9, the name, telephone number, e-mail address of the signage center shown in FIG. 1, the name of the CATV company, the telephone number, the e-mail address, etc. are notified. Various destination names, telephone numbers, and e-mail addresses are registered in the notification destination DB 134.

  In addition, by managing the address of each notification destination, for example, it is possible to identify a collaborator who lives near the security camera that detected the person to be detected. When the notification destination is a mobile phone terminal, as described above, the video management server 1 receives provision of information indicating the current position acquired by the GPS function of the mobile phone terminal, and the notification destination DB 134 By managing, it is possible to grasp the current location of each notification destination. In this case, for example, a message that calls attention to a notification destination located in the vicinity of the security camera that has detected the occurrence of an incident or accident can be notified.

[Example of data in tracking data file 135]
Next, an example of data managed by the tracking data file 135 will be described. The tracking data file 135 holds information updated by the functions of a registered image matching unit 103, a case pattern matching unit 104, and a tracking processing unit 106, which will be described in detail later. That is, in the tracking data file 135, by matching with the image data registered in the registered image DB described above, the person to be monitored (detection target person) and the animal to be monitored (detection target animal) are security cameras. When detected, the movement trajectory data of the detection target person or the detection target animal is updated.

  In the tracking data file 135, as will be described in detail later, when the occurrence of the detection target action is detected by the security camera by matching with the pattern image registered in the case / accident pattern image DB described above. The movement trajectory data of the person specified as the actor of the detection target action is updated. FIG. 10 is a diagram for explaining an example of data managed in the tracking data file 135 of this embodiment. In the tracking data file 135 of this embodiment, as shown in FIG. 10, registration IDs of detected registered image data and pattern image data, and identification IDs of security cameras that detected images corresponding to these image data, and The detection time is updated as movement trajectory data.

  In the case of the example shown in FIG. 10, the wanted crime whose registration ID of the image data is “0002” is detected at “17: 5” by the security camera whose identification ID is “3 (3)”. It has been shown. Then, in order, a wanted crime whose registration ID is “0002” is detected at “17: 8” by a security camera whose identification ID is “3 (2)”, and the identification ID is “2 (6)”. It is recorded that it was detected at “17:15” by the security camera, and detected at “17:23” by the security camera having the identification ID “4 (2)”. Thereby, the movement route of the wanted criminal can be grasped and tracked.

  In the case of the example shown in FIG. 10, an old man whose registration ID of image data is “0001” is detected at “18:07” by the security camera whose identification ID is “2 (2)”. It has been shown. Then, step by step, an elderly person with a registration ID of “0001” is detected at “18:15” with a security camera with an identification ID of “4 (1)”, and a crime prevention person with an identification ID of “2 (3)”. The fact that it was detected by the camera at “18:21” is recorded. Thereby, the movement path | route of the said elderly person can be grasped | ascertained and tracked.

  In the case of the example shown in FIG. 10, the occurrence of a traffic accident with the pattern image registration ID “J001” is detected at “18:35” by the security camera with the identification ID “2 (4)”. It has been shown. The person identified as the actor who caused the traffic accident is detected at “18:38” by the security camera having the identification ID “2 (3)”, and the identification ID is “4 (1)”. It is recorded that it was detected at “18:42” by the security camera and detected at “18:45” by the security camera having the identification ID “4 (2)”. Thereby, it is possible to grasp and track the movement route of the actor of the traffic accident. When the occurrence of an action to be detected is detected, first, the person existing in the video information is identified as the actor, and the actor's image data is retained, and the actor is tracked based on this. Will be done. Therefore, there may be a plurality of actors.

[Processing of main parts of video management server 1]
Next, the communication unit 101, the security camera video storage unit 102, the registered image matching unit 103, the incident pattern matching unit 104, the notification processing unit 105, and the tracking processing unit 106, which are the main units of the video management server 1 of this embodiment. Will be described in detail.

  First, the communication unit 101 realizes a function for the video management server 1 to transmit and receive information through the communication network 100. That is, the communication unit 101 receives a signal transmitted to the own device through the communication network 100, converts the signal into a signal that can be processed by the own device, and supplies the signal to the control unit 110. In addition, the communication unit 101 converts a signal supplied through the control unit 110 into a signal in a format to be transmitted, and transmits the signal to the communication network 100. As a result, the control unit 110 of the video management server 1 can receive and process information addressed to itself. In addition, the control unit 110 of the video management server 1 accesses a predetermined server device to refer to information, obtains necessary information, makes a call to a target destination, You can also send emails and so on.

  The security camera video storage unit 102 stores video information from various security camera systems connected to the video management server 1 through the communication network 100. That is, the video management server 1 according to this embodiment receives a video signal (video information) of a shot video transmitted from a security camera system having a security camera whose management information is registered in the camera installation position DB 131. And received through the communication unit 101. The video information from the security camera system received by the communication unit 101 is stored in the security camera video storage unit 102 through the control unit 110. The security camera video storage unit 102 stores video information for each security camera.

  The registered image matching unit 103 includes the video information stored in the security camera video storage unit 102 as described above, and the image data (reference video information) of the detection target person and the detection target animal registered in the registered image DB 132 described above. Are subjected to image recognition processing to detect a detection target person or a detection target animal in the video information. In this embodiment, data is compressed from various security camera systems using, for example, the MPEG (Moving Picture Experts Group) method, and transmitted, for example, with image data registered in units of I picture (Intra picture). Perform the matching process. If the video information transmitted from the security camera system is not compressed, for example, the image data registered at an appropriate rate, such as the rate of one for every tens of frames of video data. Comparison processing and matching processing may be performed.

  In this embodiment, since video information is transmitted from a plurality of security camera systems, the registered image matching unit 103 performs matching processing between video information and registered image data for each security camera system. Done. Note that it does not make much sense to perform matching with registered image data for video information in which no humans, animals, etc. appear and have no movement.

  Therefore, in the registered image matching unit 103, in the video information to be matched (video information for one frame), it is first determined whether or not there is a person or an animal by analyzing the video information. When there is an animal or an animal, matching processing with registered image data may be performed. Further, the matching process may not be performed when a new motion is detected by the comparison process with neighboring video information.

  Further, in the matching processing with the registered image data, it is possible to use a technique for specifying the position of the target subject person in a recent digital camera. Thus, it can be said that the matching process performed by the registered image matching unit 103 is a matching process for image recognition.

  When the registered image matching unit 103 detects the detection target person or the detection target animal in the video information from the security camera system, the registered image matching unit 103 detects the image data of the detected detection target person or the detection target animal. The registration ID of the notification destination added to is acquired, and this is notified to the notification processing unit 105.

  Based on the notified registration ID of the notification destination, the notification processing unit 105 refers to the corresponding data in the notification destination DB 134 described with reference to FIG. 9, and acquires the notification destination telephone number and email address. Based on this, a process of making a call to a notification destination or sending an e-mail is performed. In this case, based on the detected information indicating the installation position of the security camera, for example, address information can be obtained from a map database server on the communication network 100 and the address information can be notified.

  When the notification destination 105 is notified by telephone, the notification processing unit 105 forms a voice message such as “XX was found at address XX city XX town 1-chome X”. Is transmitted to the notification destination telephone terminal through the communication unit 101. In this case, the name portion such as “Mr. XXX” may be added to the corresponding data in the registered image DB 132 and the name of the individual information may be used.

  Also, in the case of notifying by e-mail, an e-mail having the same contents as the notification message described above may be created and transmitted to the notification destination. The e-mail may include a map indicating the installation position of the security camera that has captured the video in which the detection target person or the detection target animal is detected. The map can be acquired from a map database server on the communication network 100, for example. Note that a map DB may be provided in the video management server 1 in advance.

  In addition, when the registration ID of the partner registered as a cooperator is set in the notification destination of the corresponding image data in the registered image DB 132, the cooperator is called or an e-mail is transmitted. And seek protection of the person to be detected and the animal to be detected. In this case, for example, the target of protection can be reliably specified by notifying the name of the person to be detected together or by transmitting image data of the animal to be detected to the mobile phone terminal of the cooperator. To be done.

  Further, as described above, when the registered image matching unit 103 detects the detection target person or the detection target animal in the video information from the security camera, the registration ID of the image data of the detection target person or the detection target animal is detected. Then, the tracking processing unit 106 is notified of the identification ID of the security camera that detected the detection target person or the detection target animal this time and the detection time. The movement of the detection target person or the detection target animal can be tracked.

  The tracking processing unit 106 records the information from the registered image matching unit 103 in the tracking data file 135 described with reference to FIG. Based on the ID, the detection information about the image data with the same registration ID is recorded in the tracking data file 135 in association with each other as described with reference to FIG.

  When the registered image matching unit 103 detects that the detection target person or the detection target animal has moved by comparing adjacent video information, the registered image matching unit 103 considers the direction of the security camera and moves the movement direction. Is identified. Then, the registered image matching unit 103 refers to the neighboring camera information of the security camera that has detected the detection target person or the detection target animal this time in the camera installation position DB 131 described with reference to FIG. The matching processing between the video information from the image data and the image data (tracking image data) of the detection target person or the detection target animal is performed with priority over other registered image data.

  As a result, the movement of the detected detection target person or detection target animal can be tracked. Although the detection time of the detection target person or the detection target animal is not shown in FIG. 5, it can be acquired from a clock circuit provided in the video management server 1. The clock circuit is realized as an RTC (Real Time Clock) that provides the current date, current day, and current time.

  The incident pattern matching unit 104 also includes the video information stored in the security camera video storage unit 102 as described above, and the pattern image data (reference criteria) of the detection target action registered in the incident / accident pattern image DB 133 described above. The video information) is compared and matched to detect the occurrence of the detection target action in the video information. In brief, the image information from the security camera system is subjected to so-called edge detection processing, the contour of the included image portion is specified, and the shape of the specified contour portion and the shape of the contour portion of the image portion in the pattern image data It is determined whether or not an action to be detected has occurred. In this manner, the case etc. pattern matching unit 104 detects whether or not a similar image pattern exists.

  As described above, since the video information from the security camera system is data compressed by, for example, the MPEG method, for example, in the case of the registered image matching unit 103, for example, in I picture (Intra picture) units. Matching processing with registered pattern image data is performed. In addition, when the video information from the security camera system is not compressed, for example, matching with image data registered at an appropriate rate, such as the rate of one image for every tens of frames of video data. What is necessary is just to make it process.

  Further, in this embodiment, since video information is transmitted from a plurality of security camera systems, the pattern matching unit 104 for incidents, for each security camera system, also compares video information and registered pattern image data. A matching process is performed. Even in this case, there is not much point in matching video information that does not show any person or car and does not move at all with registered image data.

  For this reason, in the case matching pattern 104, whether or not there is a person, a car, a motorcycle, a bicycle, etc. by analyzing the video information in the video information to be matched (video information for one frame). If these exist, matching processing with the registered pattern image data may be performed. Further, the matching process may not be performed when there is no new movement by the comparison process with the neighboring video information.

  When the incident pattern matching unit 104 detects the occurrence of the detection target action in the video information from the security camera system, the incident pattern matching unit 104 is added to the pattern image data of the detected detection target action. The registered ID of the notification destination is acquired and notified to the notification processing unit 105.

  As described above, the notification processing unit 105 refers to the corresponding data in the notification destination DB 134 described with reference to FIG. 9 based on the notified registration ID of the notification destination, and notifies the notification destination telephone number and e-mail address. Based on this, a process of making a call to a notification destination or sending an e-mail is performed.

  When the notification destination 105 is notified by phone, the notification processing unit 105 “It seems that a traffic accident has occurred at address X city 1 × Chome X 1”, “○ city 1 ×× town 1”. A voice message such as “It seems that an injury case has occurred at address Chome X” and the like is transmitted through the communication unit 101 to the notification destination telephone terminal. In this case, the words “traffic accident” and “injury incident” may be added to the corresponding data in the incident / accident pattern image DB 133 and the “type” information may be used.

  Also, in the case of notifying by e-mail, an e-mail having the same contents as the notification message described above may be created and transmitted to the notification destination. In addition, as above-mentioned, the map which shows the installation position of the security camera which image | photographed the image | video which generate | occur | produced the detected detection target action can be included in an email. In addition, the e-mail can include an image generated by the detected action to be detected.

  In addition, when the registration ID of the other party registered as the alarm destination is set as the notification destination of the corresponding image data in the incident / accident pattern image DB 133, the alarm destination is called or an e-mail is transmitted. It is also possible to notify that attention is required due to the occurrence of an incident or accident.

  Further, as described above, when the occurrence of the detection target action is detected in the video information from the security camera, the pattern matching unit 104 of the incident or the like traces the image data of the actor of the detection target action as tracking image data. As specified and extracted. Further, the case pattern matching unit 104 notifies the tracking processing unit 106 of the registration ID of the pattern image data of the detection target action, the identification ID of the security camera that detected the occurrence of the detection target action this time, and the detection time. To do. And the movement of the actor of the detection target action can be tracked.

  The tracking processing unit 106 records the information from the registered image matching unit 103 in the tracking data file 135 described with reference to FIG. Based on the registration ID of the data, the detection information about the pattern image data with the same registration ID is recorded in the tracking data file 135 in association with each other as described with reference to FIG.

  The case matching pattern matching unit 104 mainly performs matching processing with the pattern image of the detection target action, and does not perform image recognition unlike the registered image matching unit 103. For this reason, the case pattern matching unit 104 extracts the tracking image data, which is the extracted image data of the actor, and information about the tracking image data (for example, the registered ID of the detected case and the pattern image of the accident, the case and the accident The registered image matching unit 103 is provided with the identification ID of the security camera that captured the video that was the source of detecting the occurrence of the occurrence of the camera, the shooting direction of the security camera, etc. As a result, the registered image matching unit 103 preferentially performs matching processing (here, image recognition processing) between the tracking image data from the case matching pattern matching unit 104 and the video information sequentially transmitted from the security camera system. The tracking target can be appropriately tracked.

  In this case, the registered image matching unit 103 detects the current detection target action in the camera installation position DB 131 described with reference to FIG. 6 from the identification ID of the security camera from the case matching pattern matching unit 104. Referring to the neighboring camera information of the security camera, the video information from the other security cameras in the neighborhood, and the image data (tracking image data) of the actor identified and extracted by the case matching unit 104 as described above It is possible to preferentially perform the matching process with the matching of other registered image data and the like.

  Thereby, the movement of the actor of the detected action to be detected can be tracked. Although the detection time of occurrence of the detection target action is not shown in FIG. 5, it can be acquired from a clock circuit included in the video management server 1. Also, when the registered image matching unit 103 can detect the moving direction of a person or the like corresponding to the tracking image data from the case etc. pattern matching unit 104 by using, for example, video information to be processed and video information in the vicinity thereof. Alternatively, in consideration of the detected moving direction, a security camera as a video information providing destination for which matching processing is preferentially performed may be specified.

[Summary of processing of video management server 1]
Next, processing performed in the video management server 1 described above will be described with reference to the flowcharts of FIGS. 11 and 12 are flowcharts for explaining processing performed in the video management server 1. The processing shown in FIGS. 11 and 12 is mainly performed by the communication unit 101, which is the main part of the video management server 1, in accordance with the control of the control unit 110 while the video management server 1 is powered on. This processing is performed by the image matching unit 103, the case and the like pattern matching unit 104, the notification processing unit 105, and the tracking processing unit 106.

  First, the video management server 1 receives video information from a security camera registered in its own camera installation position DB 131 through the communication unit 101 and stores it in the security camera video storage unit 102 through the control unit 110. (Step S101). Then, the control unit 110 controls the registered image matching unit 103 and the case pattern matching unit 104 to execute matching processing in each of them (step S102).

  That is, in step S102, the registered image matching unit 103 performs matching processing between the latest acquired video information (for example, I picture) and the image data registered in the registered image DB 132, and the pattern matching unit for incidents and the like. In 104, matching processing between the latest acquired video information (for example, I picture) and the pattern image data registered in the incident / accident pattern image DB 133 is performed. As will be described later, when there is tracking image data of a person who is the target of tracking, matching with the tracking image data is performed preferentially in the registered image matching unit 103.

  Then, the registered image matching unit 103 determines whether or not a person or animal whose image data is registered in the registered image DB 132 or a person or animal in the tracking image data is detected (step S103). If it is determined in step S103 that the person or animal to be detected has been detected, the registered image matching unit 103 notifies the notification processing unit 105 of the registration ID of the notification destination added to the image data. Then, the notification processing unit 105 performs a process of notifying the notification destination specified by the registration ID of the notification destination of the detection of the person or animal (step S104).

  Note that a plurality of notification destinations may be registered in the notification destination added to the image data, and the notification processing unit 105 performs notification processing to the plurality of notification destinations. In addition, the notification processing unit 105 refers to the notification destination DB 134 based on the notified registration destination registration ID, and identifies the notification destination. Depending on the type of collaborator and the like, notification messages having different contents are formed and transmitted to the respective notification destinations.

  Specifically, the police, the fire department, the security company, and the contractor are notified that the detection target person or the detection target animal has been detected by the security camera. Further, a warning message is notified to the alarm destination because the detection target person or the detection target animal is detected by the security camera. For example, it is meaningful when the person to be detected is a wanted crime. Further, a message requesting protection is notified to the cooperation partner because the detection target person or the detection target animal is detected by a nearby security camera. As described above, it is effective when the detection target person or the detection target animal is a demented elderly person or an escaped pet.

  Thereafter, the registered image matching processing unit 103 provides the necessary data (tracking data) described with reference to FIG. 10 to the tracking processing unit 106, and enters the tracking data file 135 of data that enables tracking. Is requested to be updated (step S105). As a result, the tracking processing unit 106 performs a process of updating the provided information as described with reference to FIG. In this case, the detected image data of the person to be detected or the animal to be detected is specified as a tracking image, and is preferentially matched in step S102.

  Then, after the process of step S105 and in the determination process of step S103, when it is determined that the person or animal in which the image data is registered is not detected, the pattern matching unit 104 for the case stores the pattern image data. It is determined whether or not an incident or accident that is a detection target action registered in the incident / accident pattern image DB 133 is detected (step S106). In the determination process of step S106, when it is determined that the case where the pattern image data is registered or the occurrence of an accident is not detected, the case pattern matching unit 104 notifies the control unit 110 of this, and the control unit 110 Each unit is controlled to repeat the processing from step S101.

  If it is determined in the determination process in step S106 that an incident in which pattern image data is registered or the occurrence of an accident is detected, the incident pattern matching unit 104 proceeds to the process in step S107 in FIG. The notification processing unit 105 is notified of the registration ID of the notification destination added to the image data, and the notification processing unit 105 detects the occurrence of an incident or accident at the notification destination specified by the registration ID of the notification destination. A notification process is performed (step S107).

  A plurality of notification destinations may be registered as notification destinations added to the pattern image data. For this reason, also in the processing of step S107, the notification processing unit 105 performs notification processing to the plurality of notification destinations. The notification processing unit 105 refers to the notification destination DB 134 based on the notified registration ID of the notification destination, and identifies the notification destination. The notification destination is the police, the fire department, the security company, the contractor, the alarm destination, Depending on the type of collaborator and the like, notification messages having different contents are formed and transmitted to the respective notification destinations.

  Specifically, the police, fire department, security company, and contractor are notified that an incident or accident has been detected by a security camera. In addition, a warning message is notified to the alarm destination because an incident or accident is detected by the security camera. Further, for example, a message requesting protection of an injured person is notified to the cooperation partner because the occurrence of an accident has been detected by a nearby security camera.

  After that, as described above, the incident pattern matching unit 104 identifies an actor of an action to be detected such as an incident or an accident, extracts the actor's image data as tracking image data, and holds this (Step S108). The case pattern matching unit 104 provides the tracking processing unit 106 with the necessary data (tracking data) described with reference to FIG. An update is requested (step S109). As a result, the tracking processing unit 106 performs a process of updating the provided information as described with reference to FIG.

  Thereafter, the case pattern matching unit 104 detects the tracking image data extracted in step S108 and information about the tracking image data (for example, the detected case, the registration ID of the pattern image of the accident, the occurrence of the case or the accident) The identification ID of the security camera that captured the video that is the source of the image is provided to the registered image matching unit 103 (step S110). Thereby, priority is given to the matching processing in the registered image matching unit 103 between the tracking image data and the video information sequentially transmitted from the security camera system, and the specified actor as the tracking target is appropriately tracked. To be able to do that. Then, after the process of step S110, the process from step S101 is repeated, and the process for processing the newly transmitted video information is performed.

[Effect of the embodiment]
As described above, the security camera system according to this embodiment includes video information (monitoring target video information) from a number of security camera systems installed in a predetermined area, including the security camera system already installed. Managed by the video management server 1. The video management server 1 detects the detection target person and the detection target animal by performing matching processing with the image data of the detection target person and detection target animal registered in advance and the pattern image data of the detection target action. In addition, it is possible to detect the occurrence of an action to be detected and notify this to a predetermined notification destination.

  That is, the video management server 1 registers in advance video information from a plurality of security camera systems, such as detection of a detection target person or a detection target animal, or detection of occurrence of a detection target action, with video information from a plurality of security camera systems. This can be performed based on reference video information such as image data and pattern image data.

  Therefore, it is possible to detect a predetermined event (thing) such as detection of a detection target person or a detection target animal or detection of occurrence of a detection target action in real time. Can contribute.

  In addition, if the detected person or animal to be detected moves, it can be tracked, and if the actor of the detected action to be detected is identified and this actor moves, This can be tracked. Therefore, if an incident or accident occurs, it can contribute to the solution. In this way, it is possible to track the detection target person, the detection target animal, or the actor of the detection target action, so that it is possible to detect when they are out of the monitoring area, and there is also a wanted criminal in the monitoring area. It is possible to accurately detect the case where it comes.

[Modification]
In the embodiment described above, when the detection target person or the detection target animal is detected by the registered image matching unit 103, or when the occurrence of the detection target action is detected by the pattern matching unit 104 such as an incident, Although a predetermined notification destination is notified immediately, the present invention is not limited to this.

  For example, when a detection target person or a detection target animal is detected or occurrence of a detection target action is detected, an image based on the detected video information from the security camera is displayed, for example, an external I / F 123, input / output The person in charge confirms through the monitor receiver connected through the terminal 122. When a predetermined operation is performed through the input operation unit 121 when the detection target person or the detection target animal can be confirmed in the video information or when it is confirmed that the detection target action has occurred reliably. The notification processing by the notification processing unit 105 can be started. If this is essential, it is possible to prevent adverse effects caused by erroneous detection.

  In addition, when a detection target person or a detection target animal is detected or occurrence of a detection target action is detected, an image based on the video information from the detected security camera is transmitted to the notification destination, and the notification destination You may enable it to confirm the truth of detection.

  Further, in the above-described embodiment, the shooting direction of the security camera of the security camera system is basically fixed in one direction. However, the present invention is not limited to this. Some cameras can remotely control the direction of the security camera. In this case, it is also possible to change the shooting direction of the security camera by transmitting a control signal from the video management server 1 through the communication unit 101 and the communication network 100.

[Others]
In the above-described embodiment, the registered image DB 132 and the incident / accident pattern image DB 133 of the video management server 1 realize a function as a storage unit. In addition, the registered image matching unit 103 and the case matching pattern matching unit 104 of the video management server 1 realize a function as a detection unit, and mainly a notification processing unit 105 realizes a function as an execution unit. The tracking processing unit 106, the tracking data file 135, the registered image matching unit 103, and the case matching pattern matching unit 104 of the video management server 1 cooperate to realize a function as a tracking unit.

  In the embodiment described above, a system in which the video management server 1 and various security camera systems are formed through the communication network 100 is a security camera system to which an embodiment of the present invention is applied.

  Moreover, the process demonstrated using FIG. 11, FIG. 12 is a process implement | achieved by the method of this invention. A program for executing the processing described with reference to FIGS. 11 and 12 is a program executed on a computer of a server device to which an embodiment of the present invention is applied.

  Further, as the operating entity of the video management server 1 of the above-described embodiment, in addition to various companies, a specialized NPO (NonProfit Organization) corporation that becomes the operating entity may be set up in view of its public nature. When an NPO corporation becomes the operating entity, it can be trusted at low cost by collecting donations from local residents, residents' associations, and shopping associations in the monitoring area, or by receiving support from public organizations such as national and local governments. A highly reliable security camera system can be constructed.

  DESCRIPTION OF SYMBOLS 1 ... Video management server 1, 101 ... Communication part, 102 ... Security camera image | video storage part, 103 ... Registered image matching part, 104 ... Pattern matching part, such as a case, 105 ... Notification processing part, 106 ... Tracking processing part, 110 ... Control 111, CPU, 112, ROM, 113, RAM, 114, EEPROM, 115, CPU bus, 121, input operation unit, 122, input / output terminal, 123, external I / F, 131, camera installation position DB, 132 ... registered image DB, 133 ... incident / accident pattern image DB, 134 ... notification destination DB, 135 ... tracking data file, 2 ... security camera system for stores, 3 ... security camera system for home use, 4 ... city security camera system

Claims (7)

Storage means for storing one or more reference video information for detecting the occurrence of one or more predetermined events;
Receiving means for receiving video information obtained by each of one or more security camera devices installed in the city taking a video of the surroundings and transmitted through a predetermined communication network;
The video information received through the receiving means is compared with each of the one or more reference video information stored in the storage means to detect the occurrence of the predetermined event near the installation position of the security camera device Detecting means for
An execution unit that executes processing determined according to the detected event when the detection unit detects the occurrence of one or more of the predetermined events. The server device according to claim 1,
The reference video information stored in the storage means includes one or more of video information of a detection target person, video information of a detection target animal, and pattern video information of a detection target action,
The detection means can detect one or more of the detection target person, the detection target animal, and the occurrence of the detection target action according to the reference video information in the video information received through the reception means. And
The server device according to claim 1, wherein the execution unit notifies the other party according to the event detected by the detection unit of the occurrence of the event. The server device according to claim 2,
When any of the detection target person, the detection target animal, and the occurrence of the detection target action is detected in the received video information by the detection means,
Tracking means for tracking the movement of the detection target person, the detection target animal, or the actor of the detection target action in the video information based on video information from a neighboring security camera device. The server apparatus characterized by the above-mentioned. The server device according to claim 2 or claim 3, wherein
The server apparatus characterized in that the notification destination of the occurrence of the event of the execution means of the server apparatus is a communication terminal apparatus of a user registered in advance, and a counterpart provided with means for outputting notification information. Storing one or more reference video information for detecting occurrence of one or more predetermined events in a storage means;
A receiving step of receiving, through a receiving means, video information obtained by shooting a surrounding video transmitted from each of one or more security camera devices installed in the city through a communication network;
The detecting means collates the video information received in the receiving step with each of the one or more reference video information stored in the storage means, and the predetermined event in the vicinity of the installation position of the security camera device is detected. A detection process for detecting occurrence;
A security camera image comprising: an execution step in which an effective means executes a process determined according to the detected event when the occurrence of any one or more of the predetermined events is detected in the detection step How to Use. A security camera video use program executed by a computer installed in the server device of a security camera system formed by connecting one or more security camera devices installed in a city and a server device through a communication network. ,
Storing one or more reference video information for detecting occurrence of one or more predetermined events in a storage means;
A receiving step of receiving video information transmitted through the communication network through receiving means, which is video information obtained by each of the one or more security camera devices capturing a surrounding video;
The video information received in the receiving step is collated with each of one or more reference video information stored in advance in the storage means, and the occurrence of the predetermined event near the installation position of the security camera device is detected. A detection step to detect;
When the detection step detects the occurrence of one or more of the predetermined events, an execution step is executed to execute a process determined according to the detected event. Security camera video usage program. A security camera system formed by connecting one or more security camera devices installed in a city and a server device through a communication network,
Each of the one or more security camera devices is
Shooting means that captures surrounding video and captures it as video information;
Transmission means for transmitting the video information captured by the photographing means to the server device through the communication network,
The server device
Storage means for storing one or more reference video information for detecting the occurrence of one or more predetermined events;
Receiving means for receiving the video information from each of the one or more security camera devices;
The video information received through the receiving means is compared with each of the one or more reference video information stored in the storage means, and the occurrence of the predetermined event near the installation position of the security camera device is detected. Detecting means for
A security camera system comprising: an execution unit that executes processing determined according to the detected event when the detection unit detects the occurrence of one or more of the predetermined events.

Images