JP2010022028A - Video-monitoring system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a system having duplex video recording servers or a clustered video recording server, which can continue a job smoothly, when the video recording server is switched due to failure, or the like. <P>SOLUTION: A video monitoring system includes a network camera 500, having a television camera for photographing an object and a transmission device for changing video data to IP data and outputting the data to an IP network 100; a video recording server 600, having a plurality of recording devices and receiving and recording transmitted video data; and a client terminal 800. The network camera 500, further, has an ID giving part for creating and adding an ID for each reference unit of the video data. The video recording server 600 stores the ID and the video data while associating them to each other; and when a reproduction request which specifies a specific ID is received from the client terminal 800, the server reads the video data corresponding to the ID and transmits the data to the client terminal; and by repeating this processing, continuous data are reproduced. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a video recording / reproducing system and a recording / reproducing method, and more particularly to a video recording / reproducing system and a recording / reproducing method for transmitting, recording, and browsing a video over the Internet or a local area network.

  In a video system field (hereinafter referred to as a video system) such as a video monitoring system or an image distribution system, a video transmission device that transmits video data using an IP (Internet Protocol) network such as the Internet or a LAN (Local Area Network) as a transmission path. Needs are expanding rapidly. In such a video system, a combination of a camera and a video transmission device or a network camera (referred to as a camera in which the camera and the video transmission device are integrated) is used, and video data from the camera is directly transmitted from the network. Recording recorders (hereinafter referred to as video recording servers) are becoming mainstream. Note that the above-described video transmission apparatus compresses video using, for example, MPEG2 (Moving Picture Expert Group Phase) 2 or MPEG4 and transmits video stream data (hereinafter referred to as video data) in order to transmit video captured by a camera to a network. Device).

  Thus, in conventional cameras, video data in an analog format such as the NTSC standard, which is a standard system of television (TV), is transmitted using a coaxial cable, but is attenuated during transmission because it is transmitted as an analog signal. In other words, the image quality deteriorates due to deterioration due to noise. Moreover, since it is necessary to lay a coaxial cable for each camera, the cable laying cost becomes enormous as it is remote, and the remote monitoring system that is put into practical use is very limited.

  However, the video data from the network cameras and video transmission devices described above already have IP networks such as the Internet and local area LANs already installed widely in buildings and areas, and installation work such as newly laying IP networks becomes unnecessary. The video system can be constructed very easily. As a result, for example, if it is connected to the Internet, it is possible to view local videos very easily, not only in Japan but also from overseas. In addition, even when newly laying, for example, if a wireless LAN is used, it is not necessary to install a transmission path to the camera by simply installing a relay station, and the laying cost can be greatly reduced.

  In addition, since relay transmission equipment can use general PC peripheral equipment as it is, it is much cheaper than an analog camera. In addition, since it is an IP network system, it is easy to design a system in cooperation with various types of network components (sensors, personal computers, servers, etc.). For example, in an entrance / exit management system in a building using an IC card, a face photo image of an entrance / exit person is taken with a network camera, and a frame image of a person with a face photo is stored together with an entrance / exit management database. This makes it possible to construct an advanced security network system that links entry / exit information and video.

  A video system using an IP network has many advantages as described above, but is less reliable than an analog camera system because the underlying technology and components are personal computer peripherals. Unlike the NTSC system, the underlying IP network is an asynchronous communication system that assumes data loss, for example, discarding packets when video data is packetized and sent. For this reason, the video may be temporarily interrupted due to the occurrence of a computer virus or sudden network load, or in the worst case, the entire video system may be down.

  In addition, personal computer peripherals are unstable even at the device level compared to analog camera systems. High-function and complex OS (Operating System) such as Windows (registered trademark of Microsoft) and Linux (registered trademark) are adopted as video devices such as cameras and recording devices become more sophisticated and complicated. As a result, the video system has been seen to freeze. For this reason, when a high degree of reliability is required even in a video system, it is necessary to take measures such as duplication or clustering of networks and devices like a server in a data center.

  Note that “duplexing” here refers to preparing another server for backup that performs exactly the same function as the server that is normally operated on the main server, and switching to the backup server when a failure occurs on the main server The system is continuously operated even after a failure occurs. However, the backup server must always mirror the main server. Here, mirroring means that the backup server is written with the same data as the main server and operates in the same manner.

  In addition, “clustering” is a server configuration method for distributing a load concentrated on one server, using a plurality of node servers (servers that perform actual processing of node servers) and a load distribution device. This means configuring a single virtual server. That is, a plurality of node servers form a cluster. In clustering, when a failure occurs, the failed server is disconnected, and the processing of the failed server is shared by other multiple servers, and the system is continuously operated even after the failure occurs. In order to apply clustering, individual node servers must have the same function and must be mirrored to each other. Moreover, each job (job) is independent (each session is also called independent) like a Web server, and its use is limited to a service that does not require state management between the client and the server.

  In a Web server where duplication technology or clustering technology is used as standard, each job is basically independent of operation, and it is not necessary to perform state management between a specific client and server. Even when a failover (fail over) occurs, the client can continue browsing (browsing browsing) smoothly without being affected by the failover. Here, failover is, for example, operating two servers at the same time, operating one as the main server, and when the main server becomes inoperable, the second server takes over the processing as an immediate backup. Say that.

  Further, a conventionally known multiple media recording apparatus (see, for example, Patent Document 1) describes a plurality of recording devices and a backup device. This technique is a recording required for event recording or the like. It shows a technology that makes it possible to create backups without compromising high-speed response.

JP-A-11-250578 (page 6-8, FIG. 1)

  In the network video distribution system described above, the communication mode between the client and the distribution side server is typically a method of managing the client on the distribution side. In such a system, when the server has to be switched for some reason such as failure or overload, information for managing the reproduction state is lost. As a result, it was not possible to continue from the situation before the disconnection, and there was no method other than obtaining the information from the initial state again, and it was impossible to continuously browse the video immediately before the occurrence of the failure.

  In addition, in a system using a duplication technology or a clustering technology, a mismatch occurs in link information regarding a database, for example, recording position information of a recorded video, with a plurality of servers, and a video having the same recording capacity. Even if it is a recording server, even if it is the same image according to the usage state until then, the recording position is usually completely different. For this reason, when the video recording server is switched due to an accident or the like, the link information with the database indicates a completely irrelevant video. Therefore, when there is a mismatch in the link information related to the database among a plurality of video recording servers and the video recording server is switched due to a failure or the like, the job cannot be taken over smoothly between a specific client and the server. There was a problem. For this reason, in order to build a highly reliable system, the database pair must be unitized and duplicated or clustered between multiple recording servers, the system becomes very large, and the price and operational aspects are Of course, there was a big problem with the scalability of the system.

  An object of the present invention is to provide a video recording / reproducing system and a recording / reproducing method capable of equivalently accessing a plurality of video recording servers.

  Another object of the present invention is a video recording / reproducing system in which link information related to a recorded database can be used correctly even if video is recorded in a distributed manner on a plurality of video recording servers for duplication or clustering. It is another object of the present invention to provide a recording / reproducing method.

  Another object of the present invention is to provide a video recording / reproducing system and a recording / reproducing method capable of minimizing system resources of a video recording server when link information is recorded in a database.

  Still another object of the present invention is to provide a video recording / reproducing system and a recording / reproducing method in which link information related to a database to be recorded can be used without being affected by time fluctuations.

  Still another object of the present invention is to provide a video recording / reproducing system and a recording / reproducing method capable of integrally generating video data and link information.

  The video recording / playback system of the present invention includes a video data input unit, an ID adding unit that generates an ID to be added to the video data from the video data input unit, and a plurality of video recording devices that respectively record the video data. And an output unit for outputting the video data recorded in the plurality of video recording devices, each of the plurality of video recording devices having an ID management table and a video data storage area, and an ID of the ID management table And the video data in the video data storage area are associated with each other by the recording position information of the video data.

  The video recording / reproducing system of the present invention further includes a load distribution device and an output terminal device connected to the output unit, and the load distribution device grasps a load factor of each of the plurality of video recording devices. Then, in response to the video data request from the output terminal device, the video data request is transferred to any one of the predetermined video recording devices according to the load factor of each of the video recording devices. Configured as follows.

  Furthermore, the video recording / playback method of the present invention generates a plurality of video data, adds a distinguishable ID to each of the plurality of video data, and records the video data to which the ID is added into a plurality of video recordings. When recording to a device, each of the plurality of video recording devices has an ID management table and a video data storage area, and the ID of the ID management table and the video data of the video data storage area are A predetermined video data is reproduced from the video data stored in association with the recording position information and recorded in the plurality of video recording devices based on the ID in response to a video reproduction request.

  According to the present invention, in a video system, a video recording / playback system and a recording system that achieves high reliability by realizing dual or clustering of functions for recording, browsing, and playing back video data captured by a camera. There is an effect that a reproduction method can be provided. That is, it is possible to realize a video recording / reproducing system and a recording / reproducing method capable of equivalently accessing a plurality of video recording servers. In addition, according to the present invention, a video recording / reproducing system and recording capable of seamlessly switching between a live video from a camera and a recorded video of a video recording server while minimizing system resources such as CPU, memory, and bus bandwidth. A reproduction method can be provided.

  Also, by sharing the ID added to the recorded video data and the ID added to the sensor information, the linkage processing between the video data and the sensor information can be facilitated, and system resources can be minimized. It is.

  Also, for duplication and clustering, a video recording / playback system and recording / playback method are provided that can correctly use link information related to the recorded database even if video is recorded in a distributed manner on a plurality of video recording servers. can do.

  In addition, the video system of the present invention has the effect of realizing a recording / playback system and a recording / playback method that can easily expand new functions, such as adding a video recording server or adding various sensor information such as recognition and detection. .

  Furthermore, it is possible to provide a video recording / reproducing system and a recording / reproducing method in which link information related to a database to be recorded can be used without being affected by time variation. In addition, by providing the given link information (ID) with order and continuity, it is possible to determine whether or not a desired portion of the recorded video is recorded at high speed without checking the entire recorded data. It is possible to provide a recording / reproducing system and a recording / reproducing method that can be distinguished.

  Further, by adopting a configuration in which the video data and the link information (ID) are generated integrally, it is possible to provide a video recording / reproducing apparatus and recording / reproducing method with high reliability and low price.

It is a figure which shows schematic structure of one Example of this invention. It is a figure which shows the specific structure of one Example of this invention shown in FIG. It is a figure for demonstrating another Example of this invention. It is a figure for demonstrating schematic structure of other one Example of this invention. It is a figure for demonstrating schematic structure of other one Example of this invention. It is a figure for demonstrating schematic structure of other one Example of this invention. It is a figure for demonstrating the management table of the data recorded on the video recording device of one Example of this invention. It is a figure which shows the display screen of the client terminal used for this invention.

  FIG. 1 is a video system diagram for explaining an embodiment of the present invention. In FIG. 1, reference numeral 100 denotes an IP network (hereinafter simply referred to as a network.) That performs communication using a TCP (transmission control protocol) / IP protocol such as a wired LAN, a wireless LAN, or the Internet. The network may be a transmission path. ), 200 is a video transmission device, and 300 is an image pickup device for photographing a subject, for example, a television camera. Reference numeral 400 denotes a sensor. For example, in an intruder monitoring system, an infrared sensor that detects the presence or absence of an intruder, or a water level sensor that detects the water level of a dam. Note that the video transmission device 200 is a device that converts the outputs of the camera 300 and the sensor 400 into IP data and outputs the IP data to the network 100. That is, the video output of the camera 300 is compressed and encoded into, for example, a format such as JPEG, MPEG2 or MPEG4 and converted into IP data, and the sensor information from the sensor 400 is superimposed on the video data converted into IP data. Output to.

  Reference numeral 500 denotes a network camera. In the network camera 500, the camera 300 and the video transmission device 200 are integrally configured, and video data from the network camera 500 is directly output to the network 100 as IP data. Reference numeral 600 denotes a video recording server that records video data transmitted from the video transmission apparatus 200 or the network camera 500. 700 is a data recording device (database unit) for registering sensor information and link information (ID) for designating a video related to the sensor information when sensor information is superimposed on the video data, for example. It is. In addition, in a monitoring system or the like, the date and time taken by a timer built in the camera 300 or the network camera 500 (in some cases, externally attached) is generally attached, and these times are set as IDs. At the same time, it is stored in the data recording device 700.

  A client terminal 800 is realized by a personal computer (PC) or a dedicated hardware configuration. The client terminal 800 reproduces a live video transmitted from the video transmission device 200 or the network camera 500, reproduces a video recorded in the video recording server 600, or data recorded in the data recording device 700. Used to play. Of course, the client terminal 800 has conventionally known operations such as panning, tilting, and zooming of the camera 300 and the network camera 500 (a swiveling mechanism such as a pan head is not shown) and the operation of the sensor 400 as necessary. It has the function to control by the method.

  In FIG. 1, one representative is shown as a representative one, but generally, the video transmission apparatus 200, the camera 300, the sensor 400, or the network camera 500 has several to several tens of networks 100. It is common to configure a video system connected to the network. Also, the client terminal 800 is usually installed at a plurality of locations.

  FIG. 2 is a block diagram for explaining a specific configuration of an embodiment of the present invention. 1, the video recording server 600 includes, for example, a plurality of clustered video recording devices 600-1, 600-2,... 600-n. Reference numeral 610 denotes an ID (Identification) (also referred to as link information) management table, and 620 denotes a video data recording area. The ID management table 610 and the video data recording area 620 are schematically shown. In the video recording apparatus 600-1, a recording medium such as a hard disk (HD), an optical desk, or a DVD (Digital Versatile Disc) is used. It means the recording area recorded above. Further, the ID management table 610 may be stored in a semiconductor memory or the like when a high-speed search is required. Reference numeral 630 denotes a load balancer, which will be described in detail later. 700-1 is a data recording apparatus, but is not always necessary, and is provided when time information or sensor information as described later needs to be accumulated. Note that the data recording device 700-1 can also be used as one of the video recording devices 600-n, and it is not particularly necessary to physically distinguish them. The client terminal 800 includes an input unit 810, a decryption unit 820, and display means 830. Reference numeral 900 denotes an ID assigning unit, which will be described in detail later. In addition, the same code | symbol is attached | subjected to the same thing as FIG. In FIG. 2, the network 100 is omitted, but the transmission path between the video transmission apparatus 200 and the video recording server 600 and the transmission path between the client terminal 800 and the video recording server 600 are the network 100. It corresponds to. Details of the ID will be described later.

  Next, the operation of the system shown in FIG. 2 will be described. First, video data recording will be described. Video data output from the video transmission device 200 is input to the ID assigning unit 900. Although various modifications are possible for the installation place of the ID assigning unit 900, for the sake of convenience, the case where the ID assigning unit 900 is located between the video transmission apparatus 200 and the transmission path 100 will be described. The ID assigning unit 900 inserts an ID (or ID information) for each frame of the TV, for example, a unit for referring to the video (or a unit for reproducing) into the video data input from the video transmission apparatus 200. Output to the network 100.

  The video data to which the ID is assigned is transmitted to and recorded on all of the plurality of recording devices 600-1, 600-2,... 600-n. For example, in the video recording apparatus 600-1, which is a node server, video data is recorded in the video data recording area 620 on the hard disk, for example, inside the server, and recording position information on the determined area and the assigned ID Are associated with each other and recorded in the ID management table 610.

  The relationship between the ID management table 610 and the video data recording area 620 will be described in more detail with reference to FIG. FIG. 7A shows the ID management table 610, and FIG. 7B shows the recording contents of the video data recording area 620. FIG. When the video data is input to the video recording device 600-1, as shown in FIG. 7B, the video data is stored in the area 621-1 on the video data recording area 620 together with the head address (position information) A1. The ID assigned to the video data and the position information A1 of the area 621-1 are recorded in the record 611-1 on the ID management table 610 as shown in FIG. Stored in). That is, the ID (for example, displayed by # 101) and the position information A1 are stored in the record 611-1.

  Next, when new video data is input, it is stored in the area 621-2 on the video data recording area 620 together with the head address (position information) A2. The ID assigned to the video data and the position information A2 of the area 621-2 are stored in the record 611-2 on the ID management table 610 with the ID (displayed by # 102) and the position information A2. Thereafter, each time new video data is input, it is sequentially recorded in the video data recording area 620, and the position information Am (m: 1, 2,...) And ID (#M) of the video data recording area 620 are recorded. ) (M = 101, 102,...) Are sequentially stored in the record 611.

  The above process is repeated each time video data is sent, but when the video data recording area 620 is full, the oldest data is overwritten with the latest data. That is, the recorded content is updated. Also, new data is added to the ID management table 610 accordingly. If there is a record whose reference destination is updated by overwriting the video data recording area 620, it is deleted. The video recording device 600-1 has been described in detail above, but the video data from the ID assigning unit 900 and the ID information assigned to each video data are stored in the video recording devices 600-2,... 600-n. Are input in the same manner as the video recording apparatus 600-1 and recorded in the ID management table and video data recording area of each video recording apparatus by the same method. Therefore, detailed description is omitted.

  For example, when the shooting date and time of the video is attached as time information to the video data, the time information is recorded together with the ID in the data recording device 700-1 as in FIG. That is, the time information is recorded at a location corresponding to the video data recording area. Accordingly, the ID, time information, and recording position information of the time information are the same as those shown in FIG. The recording capacities of the video recording devices 600-1, 600-2,... 600-n are large-capacity storage devices that only store captured images for about three days to one month.

  Next, reproduction of video data will be described. When a reproduction request is given from the user to the client terminal 800, the input unit 810 generates a reproduction request based on the image request command and transmits it to the clustered network video recording server 600. At this time, an ID specifying the recording location of the recorded video is attached to the image request command. The image request input (search input) and display screen of the client terminal 800 will be described with reference to FIG.

  The client terminal 800 includes an input unit 810 such as a keyboard, a decoder 820 for decoding video data, and display means 830. FIG. 8 shows an example of the search and result display screen. In FIG. 8, reference numeral 850 denotes a search and result display screen, for example. The search and result display screen 850 includes a search condition designation screen 851, a search result list display screen 852, and a playback screen 853.

  For example, the sensor 1 is designated as a search item on the search condition designation screen 851 as a reproduction request from the client terminal 800 using the input device 810 such as a keyboard. In this case, the sensor 1 is the camera No. This corresponds to 1 camera. Also, enter the search start time: January 1, 2002 00:00 to the search end time March 31, 2002 23:29, the time display order is descending, and the number of search results displayed is 25 And click the search button.

  When this search condition is input from the client terminal 800, first, the corresponding ID corresponding to the search start time and the search end time in the search condition is read from the data recording device 700-1, and the image data corresponding to the ID is reduced. Images (thumbnail images) are read from the video recording server 600 via the load balancer 630, and the reduced image 854-1, 854-2,... 854-25 (25 search results are displayed on the search result list display screen 852. Is displayed.) Here, the reduced image (thumbnail image) 854 of the video data is a pixel obtained by thinning out the number of pixels of one frame image, and is an image for displaying a large number of images in the search result list display screen 852. The reduced image 854 does not display all reduced images of a series of captured images. For example, a well-known reduced image of a frame with a change in video is selected and displayed as a representative image. The Note that the date and time of imaging is also displayed on the search result list display screen 852 as required by the user.

  Next, the operation of the load balancer 630 will be described. When the image data is read from each of the video recording devices 600-1, 600-2,... 600-n using the assigned ID, the load distribution device 630 loads the video recording server 600 that is a node server in the cluster. , The reproduction request from the client 800 is transferred to the server with the lowest load. The load distribution device 630 responds to a reproduction request from a plurality of client terminals 800 connected to the network 100 according to the load of each video recording device 600-1, 600-2,... 600-n. And switch the playback request. For example, the load distribution device 630 includes a management table for each video recording device as shown in Table 1 inside.

Table 1 shows the load factor for each video recording apparatus at a predetermined time point. Here, the load factor is, for example, the percentage of the number of playback request commands accessed from the client terminal to each video recording device. Therefore, when there is a reproduction request from the client terminal 800, the load balancer 630 refers to this table and transfers the reproduction request to the video recording apparatus with a low load factor. For example, since the load factor of the video recording apparatus 600-1 is the lowest 40%, the reproduction request is transferred to the video recording apparatus 600-1. It should be noted that a frequently used round robin method, that is, a method in which a reproduction request is sequentially transferred to each of the video recording devices 600-1, 600-2,... 600-n whenever a reproduction request occurs. Then, the load factor of all the video recording apparatuses can be made substantially uniform. However, it goes without saying that whether the operation of the load balancer 630 is the table method or the round robin method is determined by the design of the video system or the usage status of the reproduction request.

  The video recording apparatus that has received the transfer of the reproduction request, for example, the video recording apparatus 600-1, searches the ID management table 610 based on the ID assigned to the reproduction request. If there is a corresponding record, the video data is read from the hard disk and transmitted to the client 800 based on the position information Am in the video data recording area 620 written therein. The received video data is decoded by the decoder 820, appropriately reduced image processing, and displayed as a reduced image 854 on the search result list display screen 852 on the display unit 830.

  Next, the reproduction screen 853 on the display unit 830 will be described. A playback screen 853 shows a screen for playing back video recorded on the video recording server 600 (video captured and recorded by the camera). For example, when the user clicks on the reduced image 854-12 on the search result list display screen 852 with a mouse or the like, a video reproduction request is input to the load balancer 630 together with the ID attached to the reduced image 854-12. Similar to the display of the reduced image 854, a reproduction request is appropriately transferred according to the load status of the video recording devices 600-1, 6002,... 600-3, and necessary video data is reproduced. The playback screen 853 shows a case where an image of 16:30:30 00 on March 12, 2002 is being played back. The video can reproduce a still image or a moving image, and the reproduction speed can be freely changed. In addition, when the reproduction operation is continuously performed, the above processing is repeated. However, the video recording apparatus to which the reproduction request is transferred by the load distribution apparatus 630 determines the load status of each video recording apparatus at that time (see Table 1). However, since the ID is shared as described above, the corresponding video data can be output immediately. Therefore, the entire cluster, that is, load distribution processing as the video recording server 600 is realized.

  Next, an ID assigned to the above-described video data will be described. Any ID can be used as long as all IDs are guaranteed to be non-overlapping, but it is desirable that the data amount be as small as possible in view of the transmission capacity and the capacity of the recording device. On the other hand, from the viewpoint of reproduction processing, the IDs to be given are desirable as IDs because they are ordered and can be easily determined to be missing if they are continuous. For example, assume that the client 800 requests video data with a predetermined ID. Even if it is guaranteed that there is no duplication of IDs, assuming that the assigned ID is completely random, for example, in the video recording apparatus 600-1, all IDs in the ID management table 610 are received. If it is not checked, it cannot be determined whether the image data corresponding to the requested ID exists in the video recording apparatus 600-1. As described above, when the hard disk capacity of the video recording apparatus 600-1 is extremely large, the ID management table 610 cannot be accommodated on, for example, a semiconductor memory, and a part of the ID is arranged on the hard disk. The search time of the ID management table takes nearly 100 times as long as the semiconductor memory alone, and the response response when there is no video data for the requested ID is greatly reduced. If the assigned IDs are ordered and continuous, the image data can be simply checked by checking the IDs at the beginning and end of the ID management table 610 and checking whether the requested ID is within the range. Therefore, it is convenient to search for image data by giving an ID having order and continuity.

  As an example of an ID that satisfies the above-described conditions, a method that employs a natural number will be described. First, the ID assigning unit 900 incorporates, for example, a counter (or ID generation unit) for ID assignment. This counter is initialized to 1 at the time of product shipment. Each time the system is installed and operated and ID assignment is requested to the ID assigning unit 900, the internal counter is incremented by 1 and the corresponding ID is output from the output unit (not shown). In this case, however, it is necessary not to allow initialization of the counter at all during the process. Since the ID is a finite bit number, there is a possibility that duplication will occur as the ID when the counter returns to zero again. For example, in the case of TV, IDs are assigned to all frame images, and the ID is set at 30 frames per second. Even if it continues incrementing, if it is a natural number of about 40 bits, it corresponds to a frame image for about 1159, so there is no practical problem in ID assignment.

  Further, when combined with a MAC address (48 bits), which is a unique ID well known as a network device, it becomes possible to assign a unique ID to the world at the frame image level. In this case, about 88 bits are required, but the MAC address has a bit length of 24 bits (= 16,777,216) opened for each company that ships the network card. If so, it is possible to assign an ID with a shorter bit length.

  Therefore, if the application range of the system is limited, 24 bits are assigned as a product number to be assigned to the network video transmission apparatus 200 and the network camera 500, and 40 bits are assigned to the counter in the ID assigning unit 900. , About 64 bits can be used as a practically sufficient ID. Further, it goes without saying that an ID obtained by combining a camera number or a sensor number in addition to the counter output ID can be easily implemented.

  Thus, in a monitoring system or the like, the imaging time of the camera is often used as an ID of image data. A time code with this time as an ID tends to be considered to substantially satisfy the condition as an ID. For example, when image data from a plurality of cameras is supplied to the network 100, a time code between the plurality of cameras is used. A time lag problem occurs, and the time code cannot be used as an ID unless this is solved. In other words, watches used in general IT (Information Technology) equipment count the time using the vibration of the crystal oscillator, but each crystal oscillator has an error, so little by little. A time lag occurs. In the network 100, in order to cope with such a problem, the time of a device on the network 100 is adjusted using NTP (Network Time Protocol) or the like. Will be moved forward or backward.

  In addition, since the target video system handles video, the minimum unit of time shift allowed is 16 ms corresponding to one field of video (half of one frame). In general personal computers, this level of deviation occurs every 3 to 6 days, so it is difficult to ignore the time deviation. In general, since the network 100 itself is a medium that allows delay, it does not guarantee that time correction by NTP is complete. As a result, when a time code is assigned to image data as an ID, duplication or omission of ID occurs.

  Note that there is a radio timepiece as a very accurate timepiece that is currently in practical use. Since this corrects the time in units of one hour, the time problem described above can be considerably improved by installing this radio clock in the ID assigning unit and using the time code as the ID. However, although there are corrections in units of one hour, errors still occur in the meantime, so the appearance of a more accurate radio timepiece is desired. If such an accurate radio timepiece is realized, a time code can be used as an ID. Currently, the unit price around the radio clock is still expensive and difficult to adopt in an actual system. However, if these unit prices become very cheap in the future, the time code from the radio clock will be used as the ID. it can.

  FIG. 3 is a block diagram for explaining another embodiment of the present invention, and shows a water level monitoring system such as a dam for recording video data together with a sensor for acquiring water level information. In FIG. 3, the video transmission apparatus 200 includes an encoder 210, a network transmitter 220, and an ID assigning unit 900. 700-2 is a data recording device that records, for example, water level information from the sensor 400, and 710 is data in which an ID assigned corresponding to video data is associated with water level information 720 corresponding to video data. Shows the table. The client terminal 800 includes a display unit 830, a decoder 820, a search unit 840, and a search result processing unit 850. 1 and 2 are denoted by the same reference numerals. In the present embodiment, the ID assigning unit 900 is incorporated in the video transmission apparatus 200. In addition, multicast (simultaneous broadcast transmission method for a designated destination) is performed on the video recording apparatus 600-1 and the data recording apparatus 700-2 to save the transmission band of the network.

  Next, the operation of the water level monitoring system shown in FIG. 3 will be described. First, each output from the camera 300 and the sensor 400 is input to the video transmission apparatus 200. The analog video output from the camera 300 is A / D converted by the encoder 210, then compressed and encoded by the network transmitter 220 such as JPEG or MPEG, and sensor information (water level information) from the water level sensor 400. And IP data. At this time, the ID assigning unit 900 assigns the ID described above for each video reference unit (for example, one frame unit) and outputs it to the network 100.

  The video data and sensor information (water level information) to which the ID is assigned are transmitted to the video recording device 600-1 and the data recording device 700-2. In the video recording apparatus 600-1, video data is recorded in the video data recording area 620 on the hard disk, for example, in the video recording apparatus 600-1, as described with reference to FIG. The information and the assigned ID are associated with each other and recorded in the ID management table 610.

  The relationship between the video data recording area 620 and the ID management table 610 is the same as that shown in FIG. That is, when video data is input, it is stored in area 621-1 on video data recording area 620. The ID assigned to the video data (in this case, # 101) and the position information A1 on the area 621-1 are stored in the record 611-1 on the ID management table. Next, when new video data is input, the position information A2 of the area 621-2 and the ID (# 102) given to this video data are stored in the record 611-2. Thereafter, each time new video data is input, the position information Am (m = 1, 2,...) And ID (#M) (M = 101, 102,...) Of the area 620 are recorded in the record 611. Stored sequentially in -m. The above process is repeated every time video data is sent, but when the video data recording area 620 is full, the oldest data is overwritten with the latest data in order. The ID management table 610 adds new data accordingly. If there is a record whose reference destination is updated by overwriting the video data recording area 620, it is deleted. On the other hand, the video data is also transmitted to the data recording device 700-2, but is discarded without being used in the data recording device 700-2.

  In the data recording device 700-2, when sensor information (water level information) is input, it is added as a new record to the table 710 that manages the sensor information. At this time, the ID attached together is also stored together in the same record. For example, when the video data stored in the area 621-1 on the video data recording area 620 is transmitted from the video transmission device 200, the sensor information (in the first record 711-1 of the table 710 managing the sensor information) In this case, “water level 505 cm” is displayed) and ID information (in this case, “101” is displayed) is stored. Next, when new sensor information is received, sensor information (in this case, “water level 254 cm”) and ID information (in this case, 102) are stored in the record 711-2. Thereafter, each time new sensor information is received, sensor information (in this case, “water level height” information) and ID information are sequentially stored in the record 711. The above processing is repeated each time data is sent, but when the table 710 for managing sensor information runs out of space, the oldest data is discarded in order, overwritten with the latest data, and updated. In the data recording device 700-2, video data is also sent from the video transmission device 200 by multicast, but is discarded without being recorded. In the present embodiment, the data recording apparatus 700-2 has been described. However, for example, the video recording apparatus 600-2 shown in FIG. 2 can also be used. In particular, the data recording apparatus and the video recording apparatus are distinguished from each other. There is no need to do anything.

  Next, a method for searching for information recorded in the video recording device 600-1 and the data recording device 700-2 will be described. When a search instruction from the user is given to the client terminal 800, the search unit 840 transmits the command statement describing the search condition, for example, an SQL statement to the data recording device 700-2. For example, it is assumed that a search instruction from the user is an information search with a water level of 300 cm or more. In response to the search request, the data recording device 700-2 searches the table 710 for managing sensor information, collects records that match the given conditions, and returns them to the client 800 as search results. In the above case, the water level information “505 cm” stored in the record 711-1 corresponding to the information of the water level of 300 cm or higher, the ID “101” attached thereto, the water level information “325 cm”, and the ID “ID attached thereto” 104 "is sent to the search result processing means 850.

  The search result processing unit 850 separates and extracts the water level information and ID from the received search result, processes the sensor information display data so that it can be displayed in an appropriate form for the user, and displays it on the display unit 810. In addition, in relation to the water level information, for example, when a dam image is required, the ID attached to the water level information, in the above case, “101” and “104” are displayed together with the image request. It is transferred to the recording device 600-1. In the video recording apparatus 600-1, the ID management table 610 is searched based on the IDs “101” and “104” for which image requests have been made. As a result of the search, if there is a corresponding record, the video data is read from the video data recording area 620 based on the position information written therein and transmitted to the client 800. In the above case, the video data is read from the corresponding areas 621-1 and 621-4 of the video data recording area 620 from the position information A 1 and A 4 corresponding to the IDs “101” and “104” and transferred to the decoder 820. .

  In the decoder 820, the transmitted video data is decoded, processed so that it can be displayed on the display unit 830, and displayed on the display unit 830 together with the water level information from the search result processing unit 850. As shown in FIG. 8, the display format can be displayed as a reduced image 854 on the search result list display screen 852 by a well-known method, or displayed as a still image or a moving image of the captured image on the playback screen 853. You can also. In the present embodiment, the video recording device 600-1 and the data recording device 700-2 have been described as independent node servers. However, as described with reference to FIG. Needless to say, it can be operated in a simple state.

  Next, another embodiment of the present invention will be described with reference to FIGS. 4 to 6 are schematic configuration diagrams of the embodiment of the present invention classified according to where the ID assigning unit 900 is located in the video system.

  FIG. 4 is provided at the same position as the ID adding unit 900 described in FIGS. 2 and 3. When a plurality of video recording servers 600 are connected to the network 100 and the amount of data input to the data recording device 700 is large and it is inappropriate to transmit video data, This is an example in the case where it is difficult to correspond to the above.

  First, the network camera 500 transmits video data together with sensor information output from the sensor 400 to the distribution device 1000. The distribution apparatus 1000 is an apparatus that separates sensor information superimposed on video data and distributes and transmits the sensor information to the video recording server 600 and the data recording apparatus 700, respectively. The distribution device 1000 includes an ID assigning unit 900 and assigns an ID to each of the separated video data and sensor information. When there are a plurality of video recording servers 600, the video data and the ID attached thereto are transmitted by multicast as in the description of FIG. On the other hand, sensor information is transmitted to the data recording apparatus 700 using a dedicated protocol. In FIG. 4, the network camera 500 and the distribution device 1000 are physically separated, but they may be configured integrally. As described above, by providing means that combines the distribution function and the ID assigning function, the network configuration in the video system can be made flexible, and the amount of information transmitted through the transmission path can be optimized.

  FIG. 5 is a schematic configuration diagram showing a system configuration in the case of combining video recognition sensors. A video recognition sensor 1100 is connected to the network camera 500, and its output is stored in the data recording device 700. The video data is stored in a plurality of video recording servers 600. For example, in the case of an intrusion monitoring system, the video recognition sensor 1100 detects that an intruder has entered a specific place, the image recognition sensor 1100 detects it by image processing, and stores information on the presence of the intruder. It is like outputting.

  In FIG. 5, the network camera 500 is configured integrally with an ID assigning unit 900. It should be noted that the present invention is not limited to the network camera 500 and can be integrated with the camera 300. Both the video data photographed by the network camera 500 and the ID acquired from the built-in ID assigning unit 900 are transmitted to a plurality of video recording servers 600 and data recording devices 700 by multicast. The video recording server 600 records video data received by the same operation as described in FIG. On the other hand, the video recognition sensor 1100 that has received the video data and the ID performs a recognition process from the input video data and outputs the recognition result as sensor information. At this time, the ID attached to the video data is transmitted as it is attached to the sensor information, and the relationship between the data of the video recording server 600 and the data recording device 700 is maintained. As a result, the mutual processing between the image recognition and the database, which can be realized only in a single host in the past, can be realized very easily in a network environment.

  For example, consider a case where a moving object detection system is constructed with the system configuration of FIG. First, the video recognition sensor 1100 is equipped with a function for detecting a moving object. This can be realized by, for example, a conventional image processing technique. For example, the moving object can be detected by comparing the luminance level of the current image captured by the network camera 500 with the previous image and detecting the change. The video recognition sensor 1100 discriminates the moving object from the input video data, and numerically represents the position, size, and certainty of the moving object by the above-described image processing (what the moving object is experimentally determined in advance). Are stored in the data recording device 700 as sensor information.

  The client terminal 800 (omitted in FIG. 5) acquires video data from the ID included in the search result from the data recording device 700, and uses the specified position and size information to determine the area of the moving object. If it is cut out, a system for easily creating a list of moving objects can be constructed. Further, when it is intended to install a function for extracting a human face in this system, a video recognition sensor 1100 and a data recording device 700 which are newly equipped with another set of face extraction functions are prepared and moved from the client terminal 800. By searching the object database and the face image database, respectively, it is possible to easily extend to a system that searches for images that have moving objects and show faces.

  On the other hand, in the conventional single host type system, it is difficult to easily expand the system because the CPU and memory of the host computer are restricted even when the above-described facial image recognition function is added.

  FIG. 6 is a diagram for explaining a schematic configuration of a video system according to still another embodiment of the present invention. This system realizes video / sensor synchronism with the least amount of communication when the network camera 500 and the sensor 400 are juxtaposed. In other words, the ID assigning unit 900 provides a common ID to the network camera 500 and the sensor 400.

  In FIG. 6, in order for the network camera 500 and the sensor 400 to have synchronization without mutual communication, the ID assigning unit 900 is arranged at the most upstream, and the ID assigning unit 900 multicasts the ID. At this time, the ID interval is the same as the frame rate of the image captured by the network camera 500. The network camera 500 that has received the ID transmits the ID and video data to the video recording server 600 for recording. On the other hand, the sensor 400 that has received the ID similarly attaches the same ID as the ID attached to the video data to the sensor information from the sensor 400 and transmits it to the data recording device 700. By constructing such a video system, a common ID can be easily given to video data and sensor information, so that a video system that can easily record data, retrieve data, and expand the system can be realized. .

  Furthermore, if the principle of the present invention is used, a system capable of seamlessly switching between the network camera 500 and the video recording server 600 can be realized. That is, when live video is requested at the client terminal 800, the client terminal 800 is first connected to the network camera 500 via the network 100, and the network camera 500 No. The video imaged by one camera is directly displayed on the display device of the client terminal 800. On the other hand, if the client terminal 800 is the network camera 500 No. In some cases, past video captured by one camera is taken out from the video recording server 600 and the recorded video is desired to be viewed. In this case, it is necessary to switch the connection to the video recording server 600 instantaneously and shift to the process of viewing the playback video. However, in the system of the present invention, since the ID is commonly assigned, the network camera 500 No. The recorded video of one camera can be easily taken out from the video recording server 600 by referring to the ID.

  In order to realize such a system with the conventional technology, the client terminal 800 must always be connected to the video recording server 600. As a result, the system resources of the video recording server 600 are always provided to the client terminal 800 that displays the captured video from the network camera 500 for the most part of the time and requires less access to request the recorded video from the video recording server 600. System resources were wasted. In the present invention, since the video recording server 600 is connected only when the video recording server 600 is accessed, not only the use efficiency of the video recording server 600 is improved but also the video recording server is always synchronized. Since it is not necessary to operate the system 600, the design can be simplified and the system configuration can be further simplified. Thus, the system price can be greatly reduced.

  Although the present invention has been described in detail above, the present invention is not limited to the video system, the video recording / reproducing apparatus, and the recording / reproducing method described herein. Needless to say, the present invention can be widely applied to apparatuses and recording / reproducing methods.

100: Network, 200: Video transmission device, 210: Encoder, 220: Network transmitter, 300: Camera, 400: Sensor, 500: Network camera, 600 video recording server, 610: ID management table, 620: Video data Recording area, 630: load distribution device, 700: data recording device, 800: client terminal, 810: input unit, 820: decoder unit, 830: display unit, 840: search unit, 850: search result processing unit, 900: ID assigning unit, 1000: distribution device, 1100: video recognition sensor.

Claims (6)

A network camera having a television camera for photographing a subject, and a transmission device that compresses and encodes video output from the television camera and converts the encoded video data into IP data and outputs the IP data to an IP network;
A video recording server having a plurality of recording devices and receiving and recording video data transmitted from the network camera;
In a video surveillance system comprising a live video transmitted from the network camera and a client terminal that plays back video recorded on a video recording server,
The network camera further includes, for each reference unit of the video data, an ID adding unit that generates and adds an ID including information allocated to the network camera and an orderly number with respect to the reference unit,
The video recording server stores the ID and the video data in association with each other, and when receiving a reproduction request designating the ID from the client terminal, reads the video data corresponding to the ID and sends the client terminal A video monitoring system characterized in that the process of transmitting to the video is performed and the video is continuously reproduced by repeating the process. Each of the plurality of recording devices functions independently as a node server in the clustered video recording server, and records the video data to which the ID is assigned by all of the plurality of recording devices. Yes,
The video monitoring system includes a data recording device that records shooting date and time information attached to video data to which the ID is assigned together with the ID;
In response to the reproduction request from the client terminal, the load distribution device further switches the connection-destination recording device according to the load of the plurality of recording devices and transfers the reproduction request,
2. The video monitoring system according to claim 1, wherein the switched recording apparatus takes over the continuous reproduction request from the client terminal by the ID shared by the plurality of recording apparatuses. The network camera has a sensor information input unit, and the transmission device converts the sensor information from the sensor information input unit into IP data together with the encoded video data.
The video monitoring system further includes a data recording device that receives video data to which the ID is assigned and sensor information and records the ID and sensor information in association with each other.
The video recording server receives a search instruction regarding the recorded sensor information from the client terminal, searches the data recording device, and adds the sensor information that matches a search condition to the ID added to the sensor information. The video surveillance system according to claim 1, wherein the video surveillance system transmits the video playback request including the ID designation to the client. The video monitoring system receives video data to which the ID is assigned from the network camera, performs a recognition process for detecting a moving object from the input video data, and adds the recognition result to the video data. An image recognition sensor that outputs an attached ID,
A data recording device for inputting the sensor information to which the ID is assigned and recording the ID and the sensor information in association with each other;
2. The video surveillance system according to claim 1, wherein a relationship between data of the video recording server and the data recording device is maintained.   The client terminal is directly connected to the network camera without being connected to the video recording server, receives the video data to which the ID is assigned, reproduces it as the live video, and then accesses the video recording server. 2. The video according to claim 1, wherein when a past video is extracted, the video to be reproduced can be seamlessly switched from the network camera to the video recording server by extracting the past video with reference to the ID. Monitoring system.   Each of the plurality of recording devices has a plurality of records associating recording position information, which is the start address of the video data recorded in the recording area of the recording device, with the ID assigned to the video data. When there is a management table to be stored and the recording area is full and data is overwritten with new data in order from the oldest data, a record corresponding to the new data is added to the ID management table, and the reference destination is changed by the overwriting. 6. The video surveillance system according to claim 2, wherein the updated record is deleted.

Images