JP2008016896A - Monitoring apparatus and monitoring method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To efficiently search information about monitoring video images. <P>SOLUTION: Monitoring video data imaged and outputted by an imaging apparatus, and meta-data indicating information of a video image to be monitored, are acquired to execute monitoring. In this case, the meta-data are stored, meta-data coinciding with a specific filter condition are searched from stored files of the stored meta-data, and attribute information is given to the accumulated meta-data on the basis of a result of search and the attribute information is stored. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a monitoring apparatus and a monitoring method for monitoring using an imaging apparatus, and more particularly to a system using an imaging apparatus capable of generating metadata indicating information related to video data.

In a conventional monitoring system, an imaging device (video camera) and a monitoring device capable of displaying video data output by the imaging device are connected via a network, and a video is recorded by the monitoring device, so that an error can be detected. The monitoring video can be confirmed (see, for example, Patent Document 1).
JP 2003-274390 A

  By the way, when monitoring is performed with this type of monitoring system, various monitoring operations cannot be performed when it is determined whether or not an abnormality has occurred based on one preset information. For this reason, as a monitoring imaging device, information (metadata) indicating the presence or absence of moving objects is generated from the monitoring video, the information is sent to the monitoring device, and the monitoring video and metadata are used on the monitoring device side. Therefore, a device that can perform abnormality determination has been proposed and developed. When this metadata is generated together with video data and recorded by the monitoring device, the amount of data to be recorded increases in the monitoring device (recording device), and a monitoring video in a state required from the recorded data at a later date. When searching, there is a problem that it takes time and effort to search for necessary data.

  In the conventional monitoring system, for example, when saving the metadata, if the time portion changes, or if the number of video frames in the file exceeds the specified maximum number, the data is stored. The file to be saved was switched. A file is searched using the storage start time and storage end time of metadata as index information for search. This search method is suitable for starting playback from the points found by searching using the time as a key, but there is a problem that is not suitable for usage that extracts metadata that matches a specific filter. It was.

  The present invention has been made in view of such a point, and an object of the present invention is to make it possible to efficiently search for information related to a monitoring video.

  In the present invention, when monitoring video data captured and output by an imaging device and metadata indicating information related to a video to be monitored are acquired and monitored, the metadata is stored and stored. Search for metadata that matches a specific filter condition from the saved metadata file, and store the attribute information in the accumulated metadata based on the search result. It is what I did.

  By doing in this way, efficient search of metadata becomes possible using attribute information.

  According to the present invention, it is possible to efficiently search metadata using attribute information. In other words, by performing pre-processing according to the filter processing result when saving metadata, the search target can be narrowed down in advance when searching for metadata that matches the filter, and search is performed at high speed. Results will be obtained.

  Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

  FIG. 1 is a diagram showing a system configuration example of the present embodiment. Here, a plurality of network cameras 11, 12, 13,..., 19 are prepared, and the respective cameras 11-19 are connected to the servers 21, 22, 23 via the network. In the example of FIG. 1, the output of the three cameras is configured to be supplied to one server, but this is an example, and the present invention is not limited to this. The servers 21 to 23 can send data to the client terminals 31 and 32 connected via the network. As a network, a dedicated network for the monitoring system may be used, or the Internet may be used.

  Each of the network cameras 11 to 19 is an imaging device that includes an imaging unit that captures a monitoring video and outputs video data captured by the imaging unit to a network. In addition, each network camera 11 to 19 of this example includes a metadata generation unit that generates metadata that is information about the captured surveillance video, and outputs the metadata generated by the metadata generation unit. It is as. The metadata is information obtained by detecting the status of the monitoring video, such as information indicating the coordinate position of a range where there is a motion in the monitoring video and the status of the motion. Here, the metadata is generated as text data indicating the coordinate position and state. Note that all of the prepared network cameras 11 to 19 may generate metadata, but only some cameras may include a metadata generation unit.

Here, the metadata generated in the network camera will be described. The metadata is data serving as supplementary information for video data and audio data. The metadata includes, for example, the following information.
Object information (information on the result of moving object detection in the network camera. The object recognized as a moving object is an object, and the object ID, X coordinate, Y coordinate, size, etc. are entered)
・ Video time data ・ Camera pan / tilt information (orientation information)
・ Camera position information ・ Video signature information

In the following description, an example using object information and time data is described.
In the object information, the object information in the metadata is described as binary data, and the data size is minimized. Information obtained by developing object information from this metadata into a meaningful data structure such as a structure is referred to as “object information” in this document.
The alarm information is information after filtering processing based on the object information developed from the metadata. The alarm information can be obtained by analyzing metadata of a plurality of frames, determining the speed from the change of the moving object position, checking whether the moving object exceeds a certain line, or analyzing them in combination.
The alarm information includes the number of people in the video and their statistics, and can also be used as a report function.

An example of data included in the alarm information is shown below.
-Cumulative number of objects When a filter that uses the cumulative number of objects is set, the number of objects corresponding to the filter is recorded in the alarm information.
・ Object The number of objects corresponding to the filter in the frame.
-Object information corresponding to the filter Object information of the object corresponding to the filter. Contains the ID, X coordinate, Y coordinate, size, etc. of the object.

  FIG. 2 is a diagram illustrating a configuration example of a server. Although FIG. 2 shows the configuration of the server 21, other servers have the same configuration. The server 21 includes a network connection unit 91 and can perform data transmission with a network camera or a client terminal. The video data received from the network camera by the network connection unit 91 is sent to the video data buffer unit 92 and stored in the video database 94. The metadata received from the network camera by the network connection unit 91 is sent to the metadata buffer unit 93 and stored in the metadata database 95. Each of the databases 94 and 95 uses a large capacity storage means such as a hard disk drive. The metadata accumulated in the metadata database 95 is filtered by the metadata filter unit 96, and the content of the metadata is analyzed. The contents to be filtered (that is, the contents to be analyzed) are set by the filter setting unit 97. An example of filter processing will be described later. As a result of the analysis by the metadata filter unit 96, when it is necessary to switch the conditions for shooting with the camera, the rule switching unit 98 generates a command for switching the shooting rule and the like, and sends the corresponding command to the network connection unit 91. To the camera 11 or the like. Data (such as alarm information) obtained as a result of analyzing the metadata is also stored in the metadata database 95.

  2 is configured to store video data and metadata and to analyze the stored metadata, it may be configured to display video data, display analysis results, and the like. Further, when all analysis is performed on the client terminal side, the analysis process itself may be performed on the client terminal side as a configuration without the filter unit 96 or the like.

  FIG. 3 is a diagram illustrating a configuration example of a client terminal. In FIG. 3, the configuration of the client terminal 31 is shown, but the other client terminals have almost the same configuration. A client terminal, which is a terminal device that performs monitoring, includes a network connection unit 101 and can perform data transmission with a server, a network camera, or a client terminal. The video data received from the server by the network connection unit 101 is sent to the video data buffer unit 102 and stored in the video database 104. The metadata received from the network camera by the network connection unit 101 is sent to the metadata buffer unit 103 and stored in the metadata database 105. Each database 104, 105 uses a large capacity storage means such as a hard disk drive. The metadata accumulated in the metadata database 105 is filtered by the metadata filter unit 106, and the content of the metadata is analyzed. The content to be filtered (that is, the content to be analyzed) is set by the filter setting unit 107. As a result of the analysis by the metadata filter unit 106, when it is necessary to switch the conditions for shooting with the camera, the rule switching unit 108 generates a command for switching the shooting rule and the like, and the corresponding command is sent to the network connection unit 101. To the camera via the server (or directly to the camera without going through the server). The analyzed metadata is also stored in the metadata database 105.

  In addition, the client terminal includes a display unit 111 that displays a monitoring video or the like. In order to perform display on the display unit 111, the video data processing unit 109 and the metadata processing unit 110 are provided. Video data to be displayed by the video data processing unit 109 is acquired from the video database 104, and the metadata processing unit 110 Metadata synchronized with the video to be displayed is acquired from the metadata database 105. In order for the video data processing unit 109 and the metadata processing unit 110 to temporally synchronize the data to be handled, the reproduction synchronization unit 112 performs synchronization processing. When displaying live video directly supplied from a server or camera, data is sent from the video buffer unit 102 and the metadata buffer unit 103 to the display unit 111 for display.

Here, an example of filter processing in the metadata filter units 96 and 106 in the server or client terminal will be described.
Here, the filter processing information is information describing rules for generating alarm information from object information. By analyzing the metadata of multiple frames using filter processing, it is possible to create reports and output such as displaying the number of videos. For example, the following four filter types are prepared.
Appearance: An alarm occurs when an object is in a certain area.
Disappearance: An object appears in an area and is considered an alarm when it leaves the area.
Passing: Write a line and consider it as an alarm when the object exceeds it.
Capacity (number limit of objects): Count the number of people in an area, and if it exceeds a certain number, consider it an alarm.
As described above, various forms can be applied to the analysis using the metadata and the display of the analysis result depending on the purpose of monitoring.

  Next, a storage process and a search process for the stored data when the server or client terminal stores and stores the metadata in the database 95 or 105 will be described with reference to the flowcharts in FIG. The metadata is saved in a file format.

  In the present embodiment, metadata that matches a specific filter condition is searched from the stored metadata storage file, and attribute information is stored in the stored metadata based on the search result. The attribute information is saved. As specific processing for storing attribute information, here are three examples: a method of switching the storage destination according to the filter processing result, a method of having filter match information in a file, and a method of creating an index of filter match information Is shown.

First, with reference to the flowchart of FIG. 4, processing for switching the storage destination according to the filter processing result will be described.
In this case, a storage destination is designated for each filter in advance as a metadata storage format. Then, the metadata is filtered (step S11), and it is determined whether there is a filter (filter condition) that matches the filtering (step S12). If it is determined that there is a matching filter, the storage destination of the corresponding metadata is changed to the storage destination corresponding to the matched filter (step S13). Then, the metadata is stored in the set storage destination (step S14), and data is added to each storage destination based on the result of filtering the metadata. After the storing process in step S14, it is determined whether or not unprocessed data exists (step S15). If there is unprocessed data, the process returns to step S11, and there is no unprocessed data. Ends the process.

  FIG. 5 shows an example of processing when searching for stored metadata as shown in FIG. First, a filter as a search condition is designated (step S21), and it is determined whether or not a storage destination file corresponding to the filter exists (step S22). If it exists, the search target is switched to the storage destination file corresponding to the filter (step S23), and the data stored in the file is read. The storage destination can be switched by processing such as switching directories or adding identifiers to file names. Then, the metadata file to be searched is acquired, and the metadata in the file is searched (step S24). If there is no corresponding storage destination in step S22, for example, all files that match conditions such as time may be searched.

Next, an example in which filter matching information is provided in a file header or the like will be described with reference to the flowchart of FIG.
In this case, an area in which match information for each filter can be recorded is provided in the file header. Then, filter processing is performed before adding metadata to the storage file (step S31), and it is determined whether there is a matched filter (step S32). If there is a matched filter, the matching of the corresponding file is performed. The number of data is added (step S33), and the metadata is saved (step S34). Then, it is determined whether or not the number of data in the file has reached the upper limit (step S35). If the upper limit is reached, the number of matching data is cleared and a new data file is created (step S36).

  If the upper limit has not been reached in step S35, and after creating a new file in step S36, it is determined whether or not unprocessed data exists (step S37). If unprocessed data exists, Returning to step S31, if there is no unprocessed data, the process ends.

  FIG. 7 shows an example of processing when searching for stored metadata as shown in FIG. First, a filter as a search condition is designated (step S41), and filter matching information in the file is inspected (step S42). In the inspection, it is determined whether or not data matching the search target filter is included (step S43). If included, the data in the corresponding file is searched (step S44), and the file does not include Is removed from the search target and moves to the next file (step S45). The storage location of the match information is not limited to the file header but may be an external database. Further, not only at the time of saving, a new filter may be set after the fact and the matching information for the filter may be written.

  Next, a processing example for creating an index of filter match information will be described with reference to the flowchart of FIG. First, at the time of saving, a filtering process for filtering metadata is performed (step S51), it is determined whether there is a matching filter (step S52), and if there is a corresponding filter, time information (time) of the metadata is found. (Stamp), the save destination file name, and the offset position in the file are added to the metadata database as index information together with the filter ID (step S53). This process is repeated as long as there is unprocessed data (step S54).

  FIG. 9 is an example of processing when searching for stored metadata as shown in FIG. At the time of search, a filter as a search condition is designated (step S61), and search is performed from the index using the filter ID as a key (step S62). In this search, it is determined whether or not there is a matching record (step S63). If there is a matching record, the time information and storage destination information of the corresponding metadata are acquired, and all the records are acquired. (Step S64), listing and accessing the actual data (metadata) using the obtained storage destination information.

In this way, by performing preprocessing according to the filter processing result when saving metadata, it becomes possible to narrow down the search target in advance when searching for metadata that matches the filter, Metadata search results can be obtained at high speed.
Further, by recording the number of objects that match each filter, the cumulative number of data that matches the filter can be acquired without scanning the entire file.
Therefore, in the monitoring system as shown in the present embodiment, it is possible to search for the monitoring status using metadata at high speed and efficiently.

It is a block diagram which shows the system configuration example by one embodiment of this invention. It is a block diagram which shows the example of the server by one embodiment of this invention. It is a block diagram which shows the example of the terminal by one embodiment of this invention. It is a flowchart which shows the example of a preservation | save destination switching process by one embodiment of this invention. It is a flowchart which shows the example of a search by the filter conditions by one embodiment of this invention. It is a flowchart which shows the preservation | save example of the filter matching information by one embodiment of this invention. 5 is a flowchart illustrating an example of a search using filter matching information according to an embodiment of the present invention. It is the flowchart which showed the example of index creation of filter matching information by one embodiment of this invention. 5 is a flowchart illustrating an example of a search using an index according to an embodiment of the present invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 11-19 ... Network camera, 21-23 ... Server, 31, 32 ... Client terminal, 91 ... Network connection part, 92 ... Video data buffer part, 93 ... Metadata buffer part, 94 ... Video database, 95 ... Metadata database 96 ... Metadata filter unit, 97 ... Filter setting unit, 98 ... Rule switching unit, 101 ... Network connection unit, 102 ... Video data buffer unit, 103 ... Metadata buffer unit, 104 ... Video database, 105 ... Metadata database 106: Metadata filter unit, 107 ... Filter setting unit, 108 ... Rule switching unit, 109 ... Video data processing unit, 110 ... Metadata processing unit, 111 ... Display unit, 112 ... Playback synchronization unit

Claims (5)

In a monitoring device that performs monitoring using video data captured and output by a monitoring imaging device and metadata indicating information related to a monitoring target,
A metadata storage unit for storing the metadata;
A filter processing unit for searching for metadata that matches a specific filter condition from a metadata storage file stored in the metadata storage unit;
A monitoring apparatus comprising: a control unit that stores metadata stored in the metadata storage unit with attribute information based on a result searched by the filter processing unit. The monitoring device according to claim 1,
The monitoring apparatus characterized in that, as the processing based on the attribute information, the storage destination in the metadata storage unit is switched for each filter condition that matches the metadata. The monitoring device according to claim 1,
A monitoring device characterized in that, as the processing based on the attribute information, whether or not there is metadata that matches each filter condition is flagged and stored as file information of the metadata. The monitoring device according to claim 1,
As the processing based on the attribute information, the time information of the searched object, the storage file name, the in-file offset position, and the matched filter condition are stored in the metadata storage unit. In a monitoring method in which monitoring video data captured and output by an imaging device and metadata indicating information related to a video to be monitored are acquired and monitored.
Storing the metadata;
Search the saved metadata storage file for metadata that matches a specific filter condition,
A monitoring method characterized by storing attribute information in the accumulated metadata based on the retrieved result and storing the attribute information.

Images