KR20090059244A - Method and apparatus for automatic classification for dispersion control of live multi-channel image contents - Google Patents

Abstract

An automatic classification method capable of performing the high-capacity distributed processing for a live multi-channel moving picture and a system thereof are provided to classify multi-channel streaming data automatically, and assign available channels on a server. A plurality of encoding servers(30) encode moving pictures and audio data streams which are inputted after being matched with a broadcasting schedule list for a corresponding user, and then transmits live streaming data and user information through a allocated channel. A classification server(50) searches an idle channel, assigns the search idle channel to each encoding server, and then generates user classification information by linking with the received user information.

Description

Automatic classification method and system for large-scale dispersion processing of live multi-channel video {Method and Apparatus for automatic classification for dispersion control of live multi-channel image contents}

The present invention relates to an automatic classification method and a system for large-capacity distributed processing of live multi-channel video, and more particularly, the present invention allocates a channel to receive a plurality of videos coming in a live multi-channel and the user The present invention relates to an automatic classification method and system for large-capacity distributed processing of live multi-channel video that are automatically classified and stored according to each other.

In general, the exchange of information through the Internet is being used in almost all industries that require video such as Internet broadcasting stations and distance education through the development of photographing equipment that shoots videos and encoders that produce videos through them. The development of various related contents is also taking place.

With the rapid increase of the Internet users, interest in the Internet education-related video service is increasing. Currently, as the representative Internet video server, Microsoft's Windows media serve and Real Network's Real server are used.

1 is a diagram illustrating a client server-type Internet video service system according to the prior art. 1, the Internet video service system according to the related art is a basic structure for delivering a video between a media server 10 and a client 90 in real time video streaming from the encoding server 30, and the encoding server 30. ), The client 90 and the media server 20 are employed.

That is, for live webcasting, a method of encoding an image input from a camera or the like and transmitting the same to the client 90 in real time together with audio data is used. For video on demand (VOD) service, The server loads the encoded and stored data and delivers it to the client.

However, when a large number of clients are connected to one media server providing a video, a huge load is increased on the server and the network. As a technique for solving this problem, Microsoft's clustering technology is known.

2 is a system configuration diagram illustrating a clustering technique according to the prior art. Referring to FIG. 2, in the clustering method according to the related art, a plurality of media servers 21,. In other words, Internet broadcast is provided to the client 90 in multiple multichannels.

However, in recent years, the multi-content of the user concept such as Internet video lectures, UCC, etc. are increasing, the multi-content provided from a plurality of users are increasingly provided to a large number of consumers through the live internet broadcast.

In other words, the clustering method described above is possible only when each media server 2 maintains the same data. Therefore, when the video streaming data for live broadcasting enters a multi-channel, a plurality of videos are encoded in real time. It is difficult to provide live webcasting to multiple clients by encoding and transmitting.

Furthermore, since the live webcasting of the video service encodes the input image and delivers it to the client 90 in real time together with the audio data, it is pre-encoded and stored for the video on demand (VOD) service. There was a problem in that the server loaded the data and delivered it to the client.

Therefore, in order to provide a multi-channel live video input from multiple providers to multiple clients (C2C), the live video service should be available even if the incoming stream is a large number of channels (thousands or tens of thousands of channels). In order to enable such a service, an automatic classification system that automatically classifies and stores channels for multi-channel streaming data is needed.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problem, and to automatically classify incoming multichannel streaming data and assign a channel on a server for a service providing a multichannel live video to a plurality of users (C2C). It is to provide a method and system.

Another object of the present invention is to provide an automatic classification method and system for processing a VOD service by storing and classifying a multi-channel live video in real time.

The automatic classification system for distributed processing of live multichannel video according to an aspect of the present invention is an automatic classification system for distributed processing of live multichannel video. A plurality of encoding servers for receiving and encoding the input video and audio data streams matching the broadcast schedule list and transmitting the live streaming data and the user information through the assigned channel; A classification server that allocates an idle channel for receiving live streaming data of each encoding server, searches for an idle channel, allocates a channel to each encoding server, and classifies a user in association with user information received from each encoding server. And a classification server for generating information and matching and classifying the user classification information to separately classify and store the live streaming data. The method further includes reducing the number of idle channels by detecting a broadcast indication on the live streaming data received through the assigned channel. Automatic classification system for distributed multi-channel video processing.

It is preferable that the web server further provides an interface to a client connecting to view a live video service, and provides a program to the client to access a classification server.

A plurality of media servers connected to the Internet network to receive live streaming data transmitted by the encoding server under the control of the classification server and to transmit a live broadcast service to the client using the live streaming data through the Internet network; The classification server may further include a connection network independent of the encoding server, regardless of the number of clients requesting the channel reception.

The classification server is preferably characterized in that the number of idle channels is increased when detecting the end of broadcast indication on the live streaming data received through the channel allocated from the encoding server.

The automatic classification apparatus for distributed processing of live multi-channel video according to the present invention receives a broadcasting schedule list for a corresponding user who is connected to the Internet network and logging in, and matches the broadcasting schedule list with the input video and audio data streams. An automatic classification apparatus for distributed processing of a live multichannel video inputted from a plurality of encoding servers for transmitting live streaming data and user information through an allocated channel encoded by encoding the live streaming data from each of the plurality of encoding servers. An idle channel allocating unit for allocating an idle channel for receiving a channel and searching for an idle channel and allocating a channel to each of a plurality of encoding servers; And a classification processor configured to generate classification information for each user in association with user information received from each encoding server, and classify and store the live streaming data separately by matching the classification information. The number of idle channels is reduced by detecting a broadcast indication on the live streaming data received through the allocated channel.

  The classification server is preferably characterized in that the number of idle channels is increased when detecting the end of broadcast indication on the live streaming data received through the channel allocated from the encoding server.

  The classification information preferably specifies a file name and a storage folder of live streaming data according to user information or encoding server information.

  The classification information preferably designates the file name and storage folder of the live streaming data according to the scheduled date or time of the broadcast schedule list.

  It is preferable that the web server further provides an interface to a client connecting to view a live video service, and provides a program to the client to access a classification server.

  A plurality of media servers connected to the Internet network to receive live streaming data transmitted by the encoding server under the control of the classification server and to transmit a live broadcast service to the client using the live streaming data through the Internet network; The classification server may further include a connection network independent of the encoding server, regardless of the number of clients requesting the channel reception.

The automatic classification method for distributed processing of live multichannel video according to the present invention is an automatic classification method for distributed processing of live multichannel video, and a plurality of encoding servers are scheduled to broadcast to a corresponding user logging in by accessing the Internet network. A first step of receiving a list, encoding a video and audio data stream input by matching with a scheduled broadcast list, and transmitting live streaming data and user information together through an assigned channel; Allocate an idle channel for receiving live streaming data of each encoding server, search for an idle channel and allocate a channel to each encoding server, and generate user classification information in conjunction with user information received from each encoding server. And classifying and storing the live streaming data separately by matching the user classification information. And detecting a broadcast indication on the live streaming data received through the assigned channel to reduce the number of idle channels.

  A fifth step of transmitting to a client requesting a live broadcast service through an internet network by using live streaming data transmitted from an encoding server by being connected to the internet network; It is preferable to further include.

  And a fourth step of increasing the number of idle channels upon detecting the end of broadcast indication on the live streaming data received through the channel allocated from the encoding server.

According to the present invention, by automatically classifying live video of a multi-channel, there is an excellent effect of providing multi-channel streaming video input from multiple users to consumers in live or VOD form.

Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. An automatic classification system for distributed processing of live multichannel video according to the present invention will be described as a preferred embodiment when applied to an Internet service for teaching live video on the Internet.

3 is a block diagram of an automatic classification system for distributed processing of live multichannel video according to the present invention. Referring to FIG. 3, the automatic classification system for distributed processing of live multichannel video according to the present invention includes an encoding server 30, a media server 2, a web server 40, and a classification server 50. The encoding server 30 encodes video and audio data in real time and plays live streaming data through an assigned channel. The media server 2 serves to provide a multi-channel live video service to a plurality of clients 90. The media server 2 performs a distributed process to provide a multi-channel live video service to the first stage media server 20 and the plurality of clients 90 that receive the live video from the plurality of encoding servers 30. The media servers 21, ..., 24 are separated from each other. The web server 40 provides a connection between the classification server 50 and a plurality of encoding servers 30, and provides a connection interface to the client 90 connecting to watch the live video service. The classification server 50 allocates an idle channel to each encoding server 30 to deliver multiple channels of live streaming data input from the encoding server 30 to the plurality of clients 90 through the corresponding media server 2. It can play a role of classifying and storing to provide live video service of a channel.

The encoding server 30 according to the present invention having such a configuration encodes live video and audio streaming signals received and shot by a camera in a field such as a school in real time, and transmits them according to a broadcast schedule for a predetermined live video service. do. The encoding server 30 may log in a user and download a broadcasting schedule list for a predetermined live video service for the user from the web server 40 or check in real time.

Here, the broadcast schedule list may be reserved by the user in advance for the live video service of the encoding server 30 for each user. That is, the web server 40 uses user authentication to distinguish the live streaming input from the plurality of encoding servers 30, and previews the broadcasting schedule for each user in advance to distinguish the live streaming of the individual encoding server 30. Can be set.

Therefore, the present invention can efficiently distinguish the live streaming input from the plurality of encoding servers 30.

Here, the encoding server 30 encodes the video and audio data according to the broadcast schedule for the user and transmits the encoded video and audio data through the assigned channel. At this time, the encoding server 30 transmits the user information together with the streaming data.

Here, the user information may include user authentication information (user ID, user name, address, etc.) and a list of scheduled broadcasts (date of broadcast, broadcast identification number, etc.).

In addition, the encoding server 30 may insert a flag (flag) indicating the start and end of the streaming data. That is, the encoding server 30 inserts the start flag for notifying the start of the transmission of the streaming data and the end flag for notifying the end. That is, the encoding server 30 automatically distinguishes live streaming data transmitted according to the above-mentioned broadcast schedule.

Meanwhile, the classification server 50 according to the present invention receives a user login state of the encoding server 30 from the web server 40, searches for an idle channel for the corresponding user, and allocates a channel to the encoding server 30. . That is, the classification server 50 may distinguish channel allocation requests and live streaming data from a plurality of encoding servers 30 by mapping channels allocated with users.

In addition, the classification server 50 according to the present invention may detect the start flag of the streaming data input from the encoding server 30 to reduce the number of idle channels, and detect the end flag to increase the number of idle channels. .

Therefore, the classification server 50 of the automatic classification system for distributed processing of live multichannel video according to the present invention detects the number of idle channels for allocating channels for live streaming data input from each encoding server 30. Since the number of encoding servers 30 to be connected is variable, it is possible to receive live streaming data that is flexibly received.

In addition, the classification server 50 according to the present invention generates classification information for classifying live streaming data using the user information received from the encoding server 30, and the live streaming received from the encoding server 30 described above. Data can be stored according to the classification information. That is, the classification server 50 generates classification information for classifying live streaming data received through a channel allocated to each user.

Hereinafter, the classification server 50 according to the present invention will be described in more detail with reference to the drawings. In the description, description of general components having a computer function will be omitted.

4 is a block diagram showing the internal structure of a classification server according to the present invention.

4, the classification server 50 according to the present invention includes an idle channel allocator and a classification processor.

The idle channel assignment unit 52 receives a user login state of the encoding server 30 from the web server 40, searches for an idle channel for the corresponding user, and allocates a channel to the encoding server 30. That is, the idle channel allocating unit 52 of the classification server 50 may distinguish channel allocation requests from the plurality of encoding servers 30 and live streaming data by mapping channels allocated with the user.

In addition, the idle channel allocation unit 52 of the classification server 50 according to the present invention detects the start flag of the streaming data input from the encoding server 30 to reduce the number of idle channels, and detect the end flag to idle. The number of channels can be increased.

Accordingly, the classification server 50 of the automatic classification system for distributed processing of live multichannel video according to the present invention sets the number of idle channels for allocating channels for live streaming data input from each encoding server 30. Since the number of encoding servers 30 that can be detected and accessed is variable, it is possible to receive live streaming data that is flexibly received.

In addition, the classification processing unit 54 of the classification server 50 according to the present invention generates classification information for classifying live streaming data using the user information received from the encoding server 30, and the encoding server 30 described above. ) Can be stored according to the classification information. That is, the classification processing unit 54 of the classification server 50 generates classification information for classifying live streaming data received through a channel allocated to each user.

5 is an exemplary diagram showing an example of the data structure of the classification information of the present invention, Figure 6 is an exemplary diagram showing another example of the data structure of the classification information of the present invention.

As shown in FIG. 5, the classification information 100 according to an exemplary embodiment of the present invention may include a file name, a storage folder, and the like of a video for storing live streaming data according to user information or an encoding server. That is, the classification information may distinguish the corresponding live streaming by the user ID 101 or the hardware address 102 (MAC address) of the encoding server. For example, when the user ID 101 of the transmitted live stream data is "guest", the storage folder name 103 for storing the live stream data is designated as "guest", and the file name 104 of the data to be stored. ) Is set to "guest.live".

As shown in FIG. 6, the classification information 110 according to another example of the present invention may be divided according to the scheduled date and time 112 or the division number 113 of the user's scheduled broadcast schedule. For example, when the scheduled date and time 112 and the segment number 113 of the corresponding broadcast schedule list of the transmitted live stream data are "20070101" and "NO1", a storage folder name 114 for storing live stream data 114 Is designated as "20070101NO1", and the file name 115 of the data to be stored is set to "20070101NO1.live".

Accordingly, the present invention enables a plurality of live streaming broadcasts by automatically classifying and storing the live streaming input from the plurality of encoding servers 30.

In addition, the classification server 50 according to the present invention is to provide the live video service in real time via the media server (2) stored live streaming data classified in the manner described above. Of course, the media server 2 may provide a multi-channel live video broadcast to a plurality of clients 90 to access.

Of course, the classification server 50 stores live streaming data classified and stored for each user in its own DB to be transmitted to the media server 2 or transmits the above-described classification information to the corresponding media server 2 together with each media server. (2) may be classified and stored with reference to this classification information.

In addition, the media server 2 may be configured with a high performance server system in consideration of the congestion of the clients 90 that are instantaneously connected, and form a network of high-capacity high-speed communication, and each media server 21, which is formed in a hierarchical manner, may be provided with a high performance server system. , 24) may provide a live video service to a client 90 to which a clustering technique is connected. That is, by operating a plurality of media server (2), by distributing the client (90) connecting to the media server 20 of the first stage to different media servers (21, ..., 24) of two different stages, namely Multiple live channels can be provided to the client 90 via multiple channels.

In addition, the classification server 50 according to the present invention is independent of the encoding server 30 regardless of the number of clients 90 that access the media server 2 and request to receive a broadcast channel for live video service. A connection network can be provided. That is, the access network in which the classification server 50 according to the present invention distributes idle channels allocated to live streaming data received from a plurality of users may be separately configured, and a plurality of clients ( A network that distributes channels for the live video service provided by 90) may also be separately configured.

Therefore, the classification server 50 according to the present invention can be easily classified to provide live streaming data input from a plurality of users to the plurality of clients 90 through the media server 2.

Here, the classification server 50 according to the present invention classifies and stores the live streaming data input from the plurality of encoding servers 30, and provides the live streams to the plurality of clients 90 through the media server 2 in real time. The streaming data can be backed up to a separate DB (not shown) according to the classification information to enable the VOD service.

Accordingly, the present invention classifies and stores live streaming data input from a plurality of encoding servers 30 to provide a plurality of clients 90 through a media server 2 in real time, as well as a VOD service. do.

Hereinafter, an automatic classification method for distributed processing of live multichannel video according to the present invention will be described.

7 is a flowchart of an automatic classification method for distributed processing of live multichannel video according to the present invention. In the description, the same reference numerals as the above-described FIGS. 1 to 5 denote the same functions.

As shown in FIG. 7, the encoding server 30 logs in a user through the web server 40 (S10). The web server 40 authenticates the login through a user authentication procedure of the accessing encoding server 30 (S11). At this time, when the web server 40 transmits the user information on the encoding server 30 connected to the classification server 50, the classification server 50 temporarily stores the user information (S12 ~ S13).

Subsequently, the classification server 50 searches for an idle channel for transmitting a live video and allocates a channel to the encoding server 30 through the web server 40 (S14 to S16).

The encoding server 30 temporarily stores the allocated channel information and starts encoding video and audio stream data input from a camera or the like (S17 to S20).

At this time, the encoding server 30 inserts the start flag into the first data of the live streaming data to be sent to the classification server 50 through the assigned channel (S20 ~ S23).

Here, the classification server 50 stores live streaming data classified and stored for each user in its own DB to transmit to the media server 2, or transmits the above-described classification information to the corresponding media server 2 together with each media server. (2) may be classified and stored with reference to this classification information.

Thereafter, the classification server 50 sets classification information according to the user information as described above, and classifies and stores the live streaming data received in accordance with the set classification information in the media server 2 (S24 to S27). The media server 2 provides the client 90 requesting the classified and stored live streaming data.

At this time, the media server 2 may be configured with a high-performance server system to form a network of high-capacity high-speed communication in consideration of the congestion of the clients 90 that are instantaneously connected, and each media server 21... It is possible to provide a live video service to the client 90 to which .24 is applied by applying a clustering technique. That is, by operating a plurality of media server (2), by distributing the client (90) connecting to the media server 20 of the first stage to different media servers (21, ..., 24) of two different stages, namely A plurality of multi-channels provide internet live broadcasting to clients 90.

Here, when the start flag inserted into the live streaming data received from the encoding server 30 is detected, the classification server 50 reduces the number of idle channels to match the actual number of idle channels (S30).

Next, when the encoding ends, the encoding server 30 inserts the ending flag into the last data of the live streaming data to be sent (S31 to S33).

Finally, when the end flag input from the encoding server 30 is detected, the classification server 50 increases the number of idle channels to match the actual number of idle channels (S34).

In addition, the classification server 50 simultaneously stores the video for the VOD service according to the classification information to enable the VOD service (S40). That is, the classification server 50 classifies and stores the live streaming data input from the plurality of encoding servers 30 and provides the live streaming data to the plurality of clients 90 through the media server 2 in real time. The service may be backed up to a separate DB (not shown) according to the classification information to enable the service.

As such, the automatic classification method for the distributed processing of the live multi-channel video according to the present invention is for live streaming data transmitted from a plurality of encoding server 30 to be connected while repeating a series of automatic classification process (S10 ~ S40) The channel may be allocated and provided to a plurality of clients 90 simultaneously accessing the media server 2 according to the classification information.

Accordingly, the automatic classification method for distributed processing of live multi-channel video according to the present invention can detect the number of idle channels for allocating channels for live streaming data input from a plurality of encoding servers 30 and access the encoding. Since the number of servers 30 is variable, it is possible to receive live streaming data that is flexibly received, and enables multiple live streaming broadcasts by automatically classifying and storing live streaming input from a plurality of encoding servers 30. .

In addition, the present invention classifies and stores the live streaming data input from the plurality of encoding servers 30 to provide the plurality of clients 90 through the media server 2 in real time, as well as VOD services. .

While the technical idea of as described above has been described in conjunction with the accompanying drawings, this is intended to illustrate the best embodiment of the present invention by way of example and not limit the invention. In addition, it will be apparent to those skilled in the art that various modifications and variations can be made without departing from the spirit of the present invention.

1 is a block diagram showing an Internet video service system of a client server method according to the prior art.

2 is a system configuration diagram showing a clustering technique according to the prior art.

3 is a block diagram of an automatic classification system for distributed processing of a live multi-channel video according to the present invention.

Figure 4 is a block diagram showing the internal structure of the classification server according to the present invention.

5 is an exemplary diagram showing an example of a data structure of classification information of the present invention.

6 is an exemplary diagram showing another example of the data structure of the classification information of the present invention.

7 is a flowchart illustrating an automatic classification method for distributed processing of live multichannel video according to the present invention.

Explanation of symbols on the main parts of the drawings

2: media server 30: encoding server

40: web server 50: classification server

52: idle channel allocation unit 54: separation processing unit

60: client

Claims (13)

In the automatic classification system for distributed processing of live multi-channel video, Receiving a broadcast schedule for the user connected to the Internet network and logging in, encoding the input video and audio data streams matching the broadcast schedule, and transmitting live streaming data and user information through the assigned channel. Multiple encoding servers; Allocating an idle channel for receiving live streaming data from each of the plurality of encoding servers, searching for an idle channel and allocating a channel to each encoding server, and receiving user information received from each of the plurality of encoding servers; And a classification server which generates user classification information in association with each other and classifies and stores the live streaming data separately by matching the user classification information. The automatic classification system for distributed processing of live multi-channel video, characterized in that the number of idle channels is reduced by detecting a broadcast indication on the live streaming data received through the assigned channel. The method according to claim 1, And a web server providing an interface to a client connecting to view a live video service and providing a program to the client to access the classification server. Classification system. The method according to claim 1, A plurality of media servers connected to the internet network to receive live streaming data transmitted by the encoding server under the control of the classification server and transmit the live streaming data to the client requesting a live broadcast service using the live streaming data through the internet network. ; More, The classification server is an automatic classification system for distributed processing of a live multichannel video having an access network independent of the encoding server, regardless of the number of clients requesting the channel reception. The method according to claim 1, The classification server, And an increase in the number of idle channels when detecting the end of broadcast indication on the live streaming data received through the channel allocated from the encoding server. Receives broadcast schedule list for the user who is connected and logged in to the internet network, encodes the input video and audio data stream matching the broadcast schedule list, and transmits live streaming data and user information through the assigned channel. In the automatic classification device for distributed processing of live multi-channel video input from multiple encoding servers, An idle channel allocation unit for allocating an idle channel for receiving live streaming data from each of the plurality of encoding servers, searching for an idle channel and allocating a channel to each of the plurality of encoding servers; And a classification processor configured to generate classification information for each user in association with user information received from each encoding server, and to classify and store the live streaming data separately by matching the classification information. Automatic classification apparatus for distributed processing of live multi-channel video, characterized in that the number of idle channels is reduced by detecting a broadcast indication on the live streaming data received through the assigned channel. The method according to claim 5, The classification server, And an idle channel count is increased when detecting a broadcast end indication on the live streaming data received through the channel allocated from the encoding server. The method according to claim 5, The classification information, An automatic classification device for distributed processing of live multichannel video, characterized by designating a file name and a storage folder of live streaming data according to user information or encoding server information. The method according to claim 5, The classification information, And a file name and a storage folder of the live streaming data according to the scheduled date or time of the broadcast schedule list. The method according to claim 5, And a web server providing an interface to a client connecting to view the live video service and providing a program to the client to access the classification server. Automatic sorting device. The method according to claim 5, A plurality of media connected to the internet network to receive live streaming data transmitted by the encoding server under the control of the classification server and to transmit a live broadcast service to the client requesting a live broadcast service using the live streaming data through the internet network; server; More, The classification server is an automatic classification device for distributed processing of a live multi-channel video having an access network independent of the encoding server irrespective of the number of clients requesting the channel reception. In the automatic classification method for distributed processing of live multichannel video, A plurality of encoding servers receive a broadcast schedule list for a corresponding user who logs in by accessing the Internet network, encodes a video and audio data stream that is input in accordance with the broadcast schedule list, and provides live streaming data and a user through an assigned channel. A first step of sending information together; Allocating an idle channel for receiving live streaming data of each encoding server, searching for an idle channel to allocate a channel to each encoding server, and user classification information in association with user information received from each encoding server. Generating and classifying and storing the live streaming data separately by matching the user classification information; And And detecting a broadcast indication on the live streaming data received through the allocated channel to reduce the number of idle channels. 3. The automatic classification method for distributed processing of a live multi-channel video. The method according to claim 11, A fifth step of connecting to the internet network to a client requesting a live broadcast service using the live streaming data transmitted by the encoding server through the internet network; Automatic classification method for distributed processing of a live multi-channel video, characterized in that it further comprises. The method according to claim 11, And a fourth step of increasing the number of idle channels when detecting the end of broadcast display on the live streaming data received through the channel allocated from the encoding server. The automatic classification method for distributed processing of live multichannel video, further comprising: .

Images