KR20110044110A - Apparatus and method for generating metadata for surveillance system - Google Patents

Abstract

PURPOSE: An apparatus and method for generating metadata for a surveillance system are provided to exchange analysis information between intelligent surveillance systems. CONSTITUTION: A data receiving unit(110) receives a service image data from a plurality of intelligence surveillance systems using a communication protocol. A data analysis unit(120) extracts data information from the service image data. A metadata generation unit(130) generates metadata by applying the extracted information to the service image data tag.

Description

Apparatus and method for generating metadata for Surveillance System}

The present invention relates to a metadata generating apparatus and method for a video surveillance system, and more particularly, to a metadata generating apparatus for exchanging analysis information between intelligent video surveillance systems for automatically analyzing an image received from a surveillance camera; It is about a method.

Metadata is structured data about data and refers to data that describes other data.

It is data that is given to content according to a predetermined rule in order to efficiently find and use information that is being searched among a large amount of information.

Here, the location and content of the content, information about the author, conditions of rights, conditions of use, usage history, etc. are recorded.

Computers usually use metadata to express data and to find data quickly.

The best example of metadata used to represent data is HTML tags.

Data structuring is a tree-like structure from top to bottom, with head or body in the HTML tag, table in the body, tr in the table, and td in the tr. It means to achieve.

Another purpose of metadata is to find data quickly, which serves as an index of information on a computer.

Databases that we use a lot can find data quickly because this metadata is well organized.

The user can easily find specific data (information) that he / she wants using a search engine using metadata.

It is the function of metadata to extract the actor's data from one scene of the film or to extract only the shots from the video of the football match.

In the former or latter case, the metadata is invisible to the person using the data. But computers understand and use the content of metadata. In other words, metadata that the machine can understand about web material or anything else is metadata.

In general, tagging of digital data is widely known as a technique for classifying or integrated data by using information about time, space, person, and thing as a tag.

Here, tags are metadata attached to digital data for quick access or retrieval of the data.

Tag information such as metadata is mainly classified into tags for space, tags for people, tags about things, tags about time, and so on. To extract them from digital data, image analysis, barcode, and RFID (Radio Frequency) Indentification).

Intelligent Video Surveillance System consists of surveillance cameras and video recording devices connected to surveillance cameras.The video analysis function automatically analyzes video and detects events in real time according to each defined rule. to be.

Users want to be alerted in real time and to deal with threats and other important events in the environment in a timely manner.

Intelligent video surveillance systems have different rules for defining events and for exchanging message rules.

Therefore, there is a need for a monitoring system that can exchange analysis information between distributed or heterogeneous systems.

The present invention provides an apparatus and method for generating a metadata for effectively indexing a large surveillance video database and efficiently performing integrated search and management between distributed or heterogeneous surveillance systems.

An object of the present invention is to generate metadata for a video surveillance system that can effectively index a large surveillance video database and efficiently manage integrated search and management among distributed or heterogeneous surveillance systems. An apparatus and method are provided.

Meta data generating apparatus for a video surveillance system according to an embodiment of the present invention for achieving the technical problem, the data receiving unit for receiving service image data from a plurality of intelligent video surveillance system using a different platform and communication protocol; A data analyzer extracting data information from the service image data; And a metadata generator for generating metadata by assigning the extracted data information as a tag to the service image data.

Meta data generation method for a video surveillance system according to an embodiment of the present invention for achieving the technical problem, a data receiving step of receiving service image data from a plurality of intelligent video surveillance system using a different platform and communication protocol ; A data analysis step of extracting data information from the service image data; And generating metadata by attaching the extracted data information to the service image data as a tag.

According to the metadata generating apparatus and method according to the present invention, it is possible to effectively index a large-scale surveillance video database, and to efficiently operate integrated search and management between distributed or heterogeneous surveillance systems.

DETAILED DESCRIPTION In order to fully understand the present invention, the operational advantages of the present invention, and the objects achieved by the practice of the present invention, reference should be made to the accompanying drawings which illustrate preferred embodiments of the present invention and the contents described in the drawings.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Like reference numerals in the drawings denote like elements.

The present invention relates to a metadata generating device for exchanging analysis information between intelligent video surveillance systems for automatically analyzing an image received from a surveillance camera.

1 is a view showing a schematic diagram of a metadata generating device according to the present invention.

The metadata generating apparatus in the present invention includes a data receiving unit 110, a data analyzing unit 120, and a metadata generating unit 130.

The data receiver 110 receives service image data from a plurality of intelligent video surveillance systems using different platforms and communication protocols.

The intelligent video surveillance system extracts service image data from a captured image according to a preset rule, and transmits the extracted service image data using the communication protocol.

The service image data includes other information according to the set rule.

The data analyzer 120 extracts data information from service image data received from the intelligent video surveillance system.

The metadata generator 130 generates metadata by attaching the data information extracted by the data analyzer 120 to the service image data as a tag.

Meta data generated by the metadata generator 130 may be compatible with the intelligent video surveillance system.

Intelligent video surveillance systems operate on different platforms and communication protocols, but the metadata is interoperable regardless of platform or communication protocol.

Meta data generating apparatus according to the present invention may further include a transmitting unit for transmitting the meta data generated by the metadata generating unit 130 to the intelligent video surveillance system.

2 is a diagram illustrating a configuration of an intelligent video surveillance system.

The intelligent video surveillance system includes a plurality of cameras 210, a plurality of image recognition processing units 220, a plurality of recording units 230, a service processing unit 240, a rule register unit 250, and an analysis information processing unit 260. .

The plurality of cameras 210 acquires an image of each photographing area.

The plurality of image recognition processing units 220 recognizes an object in the obtained image.

A plurality of recording unit 230 records the image of the recognized object.

The service processor 240 registers a service requested by the user.

The rule register 250 stores a rule for generating service image data.

The analysis information processor 260 generates the service image data from the recognized object image based on the registered service and stored rules.

The analysis information collecting unit 270 receives only the analysis information from the image recognition processor 220 and transmits the analysis information to the analysis information processing unit 260.

The situation transmitter 280 transmits service image data to generate an alarm.

In the intelligent video surveillance system, the platform of the rule register 250 and the service processor 240, the software implementation language, and the communication protocol are different from each other, and thus have problems in interworking.

The present invention relates to a metadata generating device for exchanging analysis information between distributed systems or heterogeneous intelligent video surveillance systems.

Metadata of the present invention may be generated in XML based on an Extensible Markup Language (XML) schema or may be generated using a Document Type Definition (DTD).

Hereinafter, a method of generating metadata using XML based on XML schema will be described in detail.

The structure of metadata consists of elements relating to metadata IDs, objects, cameras, syncs, physical and logical storage locations for audio and video files.

That is, the data information extracted by the data analyzer 120 may include the type of the metadata, the type of the recognized object, the type of the camera, and the type of the recognized object. And a sink type and a type of the service video data.

The audio data file may be omitted according to the output data of the surveillance camera.

An object consists of an object ID, color, size, location, type, action, comment, and sink.

The object ID is a unique key value and has a unique value automatically generated by the system.

Color is body, head, upper and lower, and there are sub-elements. The element value is RGB (Red, Green, Blue) or HSV (Hue, Ssturation, Value). ) Or a representative color, which is a dominant color.

The body blobs the head, upper body, and lower body and has a value of RGB or HSV or a color having a high occupancy rate is designated as the representative color.

Some objects do not have head, upper and lower body values (minoccur = 0). For example, a vehicle does not divide the body into heads, upper bodies, and lower bodies, but has only representative color values for the bodies.

3 is an image obtained after extracting a foreground image based on a moving object from an image received from a surveillance camera and processing a mask.

FIG. 4 is a distribution diagram when a histogram of a hue value is represented as a circular graph when a moving object is selected as a blob in the image of FIG. 3.

4A is a histogram of the image of FIG. 3A.

In the case of achromatic color, the distribution of hue values is not prominent, so S and V values or representative colors should be used.

In FIG. 4B, since a person is wearing a yellow top, the distribution of color tones is prominent, and the color tones is the value of the element.

The size is the width, height, and width of the object. There are sub-elements. The value is an integer in cm.

Depending on image processing function, performance, utility value, and camera installation environment, the width element may be omitted (minoccur = 0).

5 is an image expressing the lengths of the horizontal and vertical units in meters by calculating the size of a moving object.

To calculate the size (y) of an object, use the following equation. Since the focal length, the height of the camera, and the tilt angle are given, the size can be calculated from the coordinates of the object input from the camera.

: 카메라의 높이,

:카메라의 틸트(tilt) 각도,

: 카메라상에서 객체의 중앙 위치,

: 카메라상에서 객체의 위 위치,

: 카메라상에서 객체의 아래 위치를 나타낸다.Where f is the focal length,

Is the height of the camera,

Tilt angle of the camera,

Is the center position of the object on the camera,

Position of the object on the camera,

: Indicates the lower position of the object on the camera.

The position represents the coordinates of x, y, z of the object.

When the GPS position of the camera, the height of the camera, and the tilt angle of the camera are input at the time of installation, the height y of the object is calculated by the above equation, and the standing position y 'can be calculated.

6 is a diagram illustrating an embodiment for calculating a size and a location of an object according to an installation environment of a camera.

The above calculation is a relative calculation that calculates only the distance away from h 'and converts the GPS coordinates of the camera's position h'.

The shape can be divided into person, bicycle, motorcycle, vehicle, boat, airplane and animal based on the size and location of the object calculated above. If it cannot be distinguished, it has an undefined value.

Additions can be made depending on the surveillance area of the video surveillance system and the types of objects that appear.

An action calculates the moving speed of an object and, in the case of humans, has a value of walking (0-150 cm / s) and running (150 cm / s or more) or a numerical value of speed.

Depending on the monitoring area, the value of the reference value (150 cm / s) can be adjusted.

Annotations include codes created in metadata management or generated error codes, and there are annotation elements for objects and metadata.

The sink inputs the time the object appears and disappears in milliseconds, and the keyframe element has a storage location as a value.

That is, the sync type of the recognized object inputs the time in which the recognized object appears and disappears in the acquired image in millisecond units.

Cameras are sub-elements such as camera ID, resolution, status, and position elements.

Resolution is a unit representing the precision of an input image and represents the number of pixels.

The resolution is the number of horizontal and vertical pixels for the image, and the state is the current operating state of the camera. The state of the camera is one of active, inactive and tampering.

Tampering is the act of knowingly or unknowingly altering or covering the camera's surveillance area.

The location element is the installation place of Carrera, and the latitude, longitude, and altitude values of the GPS values are input in the outdoors, and the indoor GPS values are used in the indoors, and each value is converted to x, y, z for use do.

 The type of the service image data and the element value of the audio and video file are the address of the file and are used to access the audio and video file corresponding to the required metadata.

7 is a flowchart illustrating a metadata generation method according to the present invention.

 In operation S710, service image data is received from a plurality of intelligent video surveillance systems using different platforms and communication protocols.

Data information is extracted from the service image data (S720).

The extracted data information is assigned to the service image data as a tag to generate metadata (S730).

As described above, optimal embodiments have been disclosed in the drawings and the specification. Although specific terms have been used herein, they are used only for the purpose of describing the present invention and are not intended to limit the scope of the invention as defined in the claims or the claims. Therefore, those skilled in the art will understand that various modifications and equivalent other embodiments are possible therefrom. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS In order to better understand the drawings cited in the detailed description of the invention, a brief description of each drawing is provided.

1 is a view showing a schematic diagram of a metadata generating device for a video surveillance system according to the present invention.

2 is a diagram illustrating a configuration of an intelligent video surveillance system.

3 is an image obtained after extracting a foreground image based on a moving object from an image received from a surveillance camera and processing a mask.

FIG. 4 is a distribution diagram when a histogram of a hue value is represented as a circular graph when a moving object is selected as a blob in the image of FIG. 3.

5 is an image expressing the lengths of the horizontal and vertical units in meters by calculating the size of a moving object.

6 is a diagram illustrating an embodiment for calculating a size and a location of an object according to an installation environment of a camera.

7 is a flowchart illustrating a metadata generation method according to the present invention.

Claims (20)

A data receiver for receiving service image data from a plurality of intelligent video surveillance systems using different platforms and communication protocols; A data analyzer extracting data information from the service image data; And And a metadata generator configured to assign the extracted data information as a tag to the service image data to generate metadata. 11. The metadata generator of claim 1, further comprising: a metadata generator configured to generate metadata. The method of claim 1, wherein the intelligent video surveillance system And extracting service image data from a captured image according to a predetermined rule, and transmitting the extracted service image data by using the communication protocol. The method of claim 1, wherein the intelligent video surveillance system A plurality of cameras for acquiring an image of a shooting area; A plurality of image recognition processing units for recognizing an object in the obtained image; A plurality of recording units configured to record an image of the recognized object; A service processor that registers a service requested by a user; A rule register for storing a rule for generating service image data; And an analysis information processor configured to generate the service image data from the recognized object image based on the registered service and a stored rule. 2. The method of claim 1, wherein the metadata (Metadata) An apparatus for generating metadata for a video surveillance system, characterized in that it is generated using Extensible Markup Language (XML) or Document Type Definition (DTD). The method of claim 1, The metadata is a metadata generating device for a video surveillance system, characterized in that the interoperability between the intelligent video surveillance system. The method of claim 1, And a transmitter for transmitting the metadata to the intelligent video surveillance system. The method of claim 3, The extracted data information includes a type of the metadata, a type of the recognized object, a type of the camera, a sink type of the recognized object, and a type of the service image data. An apparatus for generating metadata for a video surveillance system. The method of claim 7, wherein the recognized object type is And a metadata (ID), a color, a size, a location, a type, a behavior, and an annotation of the object. The method of claim 7, wherein the camera type Metadata generation apparatus for a video surveillance system comprising the ID (ID), resolution (Resoulation), status and location of the camera. The sync type of the recognized object according to claim 7, wherein the sink type of the recognized object is And a time in which the recognized object appears and disappears in the acquired image in millisecond units.  A data receiving step of receiving service image data from a plurality of intelligent video surveillance systems using different platforms and communication protocols; A data analysis step of extracting data information from the service image data; And And generating metadata by attaching the extracted data information to the service image data as a tag. 2. The metadata generation method of claim 1, further comprising: generating metadata. The system of claim 11, wherein the intelligent video surveillance system And extracting service image data from a captured image according to a predetermined rule, and transmitting the extracted service image data using the communication protocol. The method of claim 11, wherein the intelligent video surveillance system A plurality of cameras for acquiring an image of a shooting area; A plurality of image recognition processing units for recognizing an object in the obtained image; A plurality of recording units configured to record an image of the recognized object; A service processor that registers a service requested by a user; A rule register for storing a rule for generating service image data; And an analysis information processor configured to generate the service image data from the recognized object image based on the registered service and a stored rule. The method of claim 11, wherein the metadata (Metadata) Method for generating metadata for a video surveillance system, characterized in that the generation using XML (Extensible Markup Language) or DTD (Document Type Definition). The method of claim 11, The metadata is a metadata generation method for a video surveillance system, characterized in that mutually compatible between the intelligent video surveillance system. The method of claim 11, Transmitting the metadata to the intelligent video surveillance system; Metadata generation method for a video surveillance system, characterized in that it further comprises. The method of claim 13, The extracted data information includes a type of the metadata, a type of the recognized object, a type of the camera, a sink type of the recognized object, and a type of the service image data. Metadata generation method for a video surveillance system characterized in that. 18. The method of claim 17, wherein the recognized type of object is And a metadata (ID), a color, a size, a location, a type, an action, and an annotation of the object. The method of claim 17, wherein the camera type Meta data generation method for a video surveillance system comprising the ID (ID), resolution (Resoulation), status and location of the camera. 18. The method of claim 17, wherein the sink type of the recognized object is And a time in which the recognized object appears and disappears in the acquired image in millisecond units.

Images