CN102594773A

CN102594773A - Method and system for realizing data acquisition

Info

Publication number: CN102594773A
Application number: CN2011100041811A
Authority: CN
Inventors: 赵宇; 阳彦字; 王芳; 孙健; 陈翼; 李加周; 阮亚平
Original assignee: ZTE Corp
Current assignee: ZTE Corp
Priority date: 2011-01-10
Filing date: 2011-01-10
Publication date: 2012-07-18
Anticipated expiration: 2031-01-10
Also published as: CN102594773B; WO2012094992A1

Abstract

The invention discloses a method and a system for realizing the data acquisition, both of which can include, in the segmentation of a TS (Transport Stream) code stream, index information used for identifying media segmentation information, and can also acquire frames at the position of an RAP (Random Access Point) to be played in a trick mode or during the switching of a file according to the media segmentation information included in the index information. The method and the system of the invention can achieve the data acquisition technology, during fast forward/fast backward or the switching of a media stream, a client can quickly locate the start and the stop positions of the RAP to be played, and only corresponding frames are downloaded for playback. Moreover, the client can also smoothly switch to such play modes as normal play, time shifting or the like. Thus data downloaded in the trick mode or during the switching of the file can be reduced effectively, and the fluency of the play and the user satisfaction can be improved significantly.

Description

Method and system for realizing data acquisition

Technical Field

The invention relates to the field of communication, in particular to a method and a system for realizing data acquisition.

Background

With the development of technology, Streaming media services based on the HTTP protocol, i.e., HTTP Streaming service, appeared. In recent years, broadcast services and mobile services have begun to merge, with different multimedia content and services being transmitted over different networks. MPEG2-TS for broadcasting is also widely used on the internet. Also, MPEG2-TS is applied to HTTP Streaming service.

Since the HTTP protocol is essentially a text transfer protocol, it is not suitable for real-time streaming. To facilitate the functions of live playback (for time-shift services) and viewing while downloading, the program content is typically physically or logically divided into many small segments, as shown in fig. 1. The client downloads one or more segments of the Media file each time according to a program Presentation Description (MPD) containing information such as an index of the Media segment, and then decodes and plays the downloaded data. However, for the fast forward and fast backward function, a common method is to download a corresponding segment according to the MPD, and then find a corresponding Random Access Point (RAP) in a segment file for playing, so that non-RAP data is also downloaded, and thus a client downloads a lot of redundant data, which seriously affects the smoothness of playing and occupies too much network bandwidth.

Disclosure of Invention

In view of the above, the main objective of the present invention is to provide a method and a system for implementing data acquisition, so as to reduce downloaded data during trick modes or file switching and improve fluency of playing.

In order to achieve the purpose, the technical scheme of the invention is realized as follows:

a method for realizing data acquisition, index information used for identifying media segment information is contained in a segment of TS code stream, the method also includes: and acquiring a frame at a Random Access Point (RAP) to be played in a trick mode or during file switching according to the media segmentation information contained in the index information.

The method for setting the index information comprises the following steps: adding an index for identifying the media segment information in each segment of the TS code stream;

the index information includes at least the following indications: whether an RAP identifier or the number of RAPs is contained;

when there is a RAP, the index information further includes an offset address for identifying each RAP point, and a length for identifying each RAP.

The index information further comprises at least one of:

total length of RAP index information;

an RAP type for identifying a type of a corresponding segment RAP;

RAP timestamp information for identifying the playing time corresponding to each RAP;

and carrying program identification PID of RAP code stream.

The process of obtaining the frame at the RAP according to the media segment information contained in the index information comprises:

downloading and analyzing media segmentation information contained in the index information;

the offset and length of the RAP to be acquired are parsed from the media segment information and a frame at the RAP is requested to be downloaded accordingly.

The trick mode includes at least one of:

fast forward/fast backward, fast forward to normal play, time shift positioning play, file switching.

A system for realizing data acquisition comprises an index information maintenance unit and a data acquisition execution unit; wherein,

the index information maintenance unit is used for setting index information used for identifying media segment information in the segments of the TS code stream and informing the data acquisition execution unit of the index condition;

and the data acquisition execution unit is used for acquiring a frame at the RAP needing to be played in a trick mode or during file switching according to the media segmentation information contained in the index information.

When the index information maintenance unit sets the index information, the index information maintenance unit is configured to: adding an index for identifying the media segment information in each segment of the TS code stream;

The index information further comprises at least one of:

total length of RAP index information;

an RAP type for identifying a type of a corresponding segment RAP;

PID of the code stream carrying RAP.

When the data acquisition execution unit acquires the frame at the RAP according to the media segment information contained in the index information, the data acquisition execution unit is configured to:

The trick mode includes at least one of:

The method and the system realize the data acquisition technology, and the client can quickly locate the starting position and the ending position of the RAP to be played when fast forwarding/fast rewinding and media stream switching, only download the corresponding frame for playing and smoothly switch to the normal playing or time shifting playing modes and the like; therefore, the data downloaded in the trick mode or the file switching can be effectively reduced, the fluency of playing is obviously improved, and the satisfaction degree of a user can be obviously improved.

Drawings

FIG. 1 is a schematic diagram of a prior art MPEG2-TS logical segment;

FIG. 2 is a schematic diagram of the structure of an MPEG2-TS logical segment and a RAP index segment according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of generating a RAP information segment according to an embodiment of the present invention;

FIG. 4 is a flow chart of an embodiment of the present invention implementing fast forward;

FIG. 5 is a flowchart of an embodiment of the present invention for fast forward normal play;

FIG. 6 is a flowchart of an embodiment of the present invention for implementing time-shift positioning playing;

FIG. 7 is a simplified flow chart for implementing data acquisition according to an embodiment of the present invention;

fig. 8 is a system diagram for implementing data acquisition according to an embodiment of the present invention.

Detailed Description

In practical applications, for a Transport Stream (TS) file in an Http Streaming scene, when a client performs fast forward/fast backward and media stream switching, it is desirable to quickly locate the start and end positions of an RAP to be played, and only download corresponding frames for playing, and smoothly switch to a normal play mode or a time shift play mode.

As shown in fig. 2, an index for identifying the media segment information may be added in front of each segment of the TS codestream. Of course, in practical application, the index may be added after the segment of each TS code stream. Fig. 3 is a schematic diagram of a RAP information segment, and the RAP information included in the index is as follows:

the total length of the RAP index information, including this field. This field may be used to discard the RAP index data during normal on-demand. This information is optional content.

Whether an RAP identifier or the number of RAPs is contained; this information is the necessary content.

RAP type: the element identifies the type of the corresponding segment RAP; this information is optional content.

Program Identification (PID): PID of the code stream carrying RAP.

Offset address of RAP: identifying an offset address for each RAP point; this information is optional content. This field must exist when there is RAP; otherwise, this field does not exist.

Length of RAP: identifying a length of a key frame at each RAP point; this information is optional content. This field must exist when there is RAP; otherwise, this field does not exist.

RAP timestamp information: identifying the playing time corresponding to each RAP; this information is optional content.

It should be noted that, in addition to the necessary contents, the RAP information may include at least one of the above contents.

Based on the index structure, the following methods such as fast forward/fast backward, positioning play, switching and the like can be performed.

A. Method for realizing fast forward and fast backward

Further, the step a includes:

a1, the client receives trick commands (such as fast forward/fast backward);

a2, determining the next RAP address (such as URL);

the client determines the segment where the next RAP is located according to the current playing time and the fast forward/fast backward command, acquires the index information from the segment and forms the address of the frame at the RAP, wherein the index information comprises a Uniform Resource Locator (URL), a basic uniform resource locator (base of segment) + byte range (byte range) and the like;

determining the interval for downloading RAP frames according to the parameters (such as multiple) of fast forward/fast backward;

a3, the client downloads the corresponding RAP frame and decodes and displays it.

B. Method for realizing fast forward and fast backward normal play

Further, the step B includes:

b1, changing the fast forward/fast backward multiple to 1 times, and then performing corresponding fast forward/fast backward operation until the downloaded RAP is an instant update access point (IDR);

b2, obtaining the offset and length of the RAP from the RAP information segment;

b3, the client requests to download all the data of the TS segment following the RAP, and changes to normal play after downloading.

C. Method for realizing time-shifting positioning request

Further, the step C includes:

c1, finding out the corresponding TS subsection according to the time-shifting on-demand time point;

c2, calculating the display time of the time-shifting on-demand point. Finding a RAP information segment URL corresponding to the segment according to MPD, and requesting to download the RAP information segment;

c3, analyzing the RAP information segment, finding the RAP with PTS larger than or equal to the on-demand point display time and the RAP type being IDR;

c4, the client requests to download all the data of the TS segment following the RAP, and changes to normal play after downloading.

D. The method for realizing media stream switching comprises the following steps:

further, the step D includes:

d1: when the network bandwidth changes, the client or the system detects the change and determines to perform media stream switching (for example, switching from a media file 1 to a media file 2 with different resolution and other properties);

d2: the client determines the switching position according to the current playing time and the segment information (such as the segment, the time information and the RAP information corresponding to the media file 2);

d3: the client requests the media segment (e.g., segment of media file 2) at the switch location, decodes and displays the data frame, and implements media adaptive switching.

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in detail with reference to the accompanying drawings and specific embodiments.

It should be noted that, in the embodiment of the present invention, MPEG2-TS is used as the media file storage format on the streaming media server, and logical segmentation is performed on the media file storage format. The MPEG2-TS logical segment is preceded by a RAP message for that segment. So that the corresponding RAP can be quickly found out when the operations such as fast forward/fast backward, positioning play and the like are carried out, and the processing capacity of the server side is improved.

In addition, in practical application, only the RAP information part needs to be simply discarded for the prior normal on-demand and live broadcasting. The first field of the RAP index information is the total length of the RAP index. According to the length, the client can easily discard the RAP index information. When the user performs the trick mode, the exchange flow of the server and the client is changed. As shown in FIG. 4, the finding of RAPs may be accelerated using the present invention. The flow shown in fig. 4 includes the following steps:

in step 401, fast forward begins.

In step 402, it is determined whether fast forwarding is complete. If so, jumping to step 410; otherwise, go to step 403.

Step 403, determining the segment where the next RAP is located according to the current playing time and the fast forward multiple.

Step 404, determine whether the RAP index information corresponding to the segment has been downloaded. If yes, go to step 406; otherwise jump to step 405.

In step 405, the location information of the TS segment is found in the MPD. The first 1024 bytes of the segment are downloaded. The RAP index information is at the beginning of the TS segment, and since the RAP index information is small, 1024 bytes are already large enough, so there is no need to record offset and size in the MPD for the RAP information.

In step 406, the client parses the downloaded RAP index information (i.e. the media segment information included in the index information, the same applies below).

Step 407, the offset and length of the RAP are obtained.

In step 408, the client requests the server to download the frame at the RAP.

In step 409, the client processes the frame at the downloaded RAP, and then jumps to step 402.

And step 410, ending.

It can be seen that, on the basis of setting the index information, the media segment information contained in the index information can be downloaded and analyzed, then the offset and the length of the RAP to be acquired are analyzed from the media segment information, and accordingly, the frame at the RAP is requested to be downloaded.

In fast forward to normal play, a process as shown in fig. 5 may be performed, the process including the steps of:

step 501, fast forward to normal play begins.

Step 502, obtain the last RAP during fast forward.

Step 503, determine whether the RAP is IDR. If yes, jumping to 507; otherwise, go to step 504.

In step 504, fast forward operation is performed after the fast forward multiple is changed to 1.

Step 505, fast forward operation.

Step 506, obtain next RAP and jump to 503.

Step 507, obtaining the offset and length of the RAP from the RAP information segment.

In step 508, the client requests to download all the data of the TS segment following the RAP.

In step 509, the next segment is converted to the normal playing flow.

And step 510, ending.

When performing time-shift positioning playing, a flow as shown in fig. 6 may be performed, assuming that the live broadcast in fig. 6 only has a video stream, i.e. only one track in the code stream. The flow shown in fig. 6 includes the following steps:

step 601, obtaining the time of time shifting positioning playing.

Step 602, judging whether the acquired time is coordinated Universal Time (UTC), if so, jumping to step 604; otherwise, go to step 603.

Step 603, converting the time of the time shift positioning into the UTC time.

Step 604, start UTC of the TS file is obtained from the MPD.

Step 605, calculate the showing time of the on-demand time point.

Step 606, traversing the starting presentation time of each segment recorded in the MPD, and calculating the segment where the playing point is located.

Step 607, acquiring the position information of the segments from the MPD, and downloading the first 1024 bytes. Since the RAP index information is at the beginning of the TS segment and 1024 bytes are already large enough because the RAP index information is small, there is no need to record the offset and size in the MPD for the RAP information.

Step 608, traverse the RAP index information, and accordingly find the RAP with RAP type IDR and PTS greater than or equal to the presentation time of the on-demand time point.

Step 609, according to the offset and size of the RAP and the URL of the TS segment, constructs the URL of all data of the segment after the RAP, and requests the server to download.

Step 610, the next segment is converted into a normal playing flow.

With reference to the foregoing embodiments, it can be seen that the operational idea of the present invention for implementing data acquisition can be represented by a flow shown in fig. 7, where the flow includes the following steps:

step 710: index information for identifying media segment information is set in the segments of the TS code stream.

Step 720: and acquiring a frame at the RAP to be played in the trick mode or during file switching according to the media segmentation information contained in the index information.

In order to ensure that the above embodiments and operation idea can be realized smoothly, the arrangement shown in fig. 8 can be performed. Referring to fig. 8, fig. 8 is a system diagram for implementing data acquisition according to an embodiment of the present invention, where the system includes an index information maintenance unit and a data acquisition execution unit that are connected to each other.

When the method is applied specifically, the index information maintenance unit can set index information used for identifying media segment information in the segments of the TS code stream and inform the data acquisition execution unit of the index condition; and acquiring the frame at the RAP required to be played in the trick mode or file switching by the data acquisition execution unit according to the media segmentation information contained in the index information.

In summary, the present invention realizes the data acquisition technology, and when the client performs fast forward/fast backward and media stream switching, the client can quickly locate the start and end positions of the RAP to be played, only download the corresponding frames for playing, and smoothly switch to the normal playing or time-shifting playing modes; therefore, the data downloaded in the trick mode or the file switching can be effectively reduced, the fluency of playing is obviously improved, and the satisfaction degree of a user can be obviously improved.

The above description is only exemplary of the present invention and should not be taken as limiting the scope of the present invention, and any modifications, equivalents, improvements, etc. that are within the spirit and principle of the present invention should be included in the present invention.

Claims

1. A method for realizing data acquisition is characterized in that index information used for identifying media segmentation information is contained in a segment of a transport stream TS (transport stream) code stream, and the method also comprises the following steps: and acquiring a frame at a Random Access Point (RAP) to be played in a trick mode or during file switching according to the media segmentation information contained in the index information.

2. The method of claim 1, wherein the index information is set by: adding an index for identifying the media segment information in each segment of the TS code stream;

3. The method of claim 2, wherein the index information further comprises at least one of:

total length of RAP index information;

an RAP type for identifying a type of a corresponding segment RAP;

and carrying program identification PID of RAP code stream.

4. The method according to any of claims 1 to 3, wherein the process of obtaining the frame at the RAP according to the media segment information contained in the index information comprises:

5. The method of claim 4, wherein the trick mode comprises at least one of:

6. A system for realizing data acquisition is characterized by comprising an index information maintenance unit and a data acquisition execution unit; wherein,

7. The system according to claim 6, wherein the index information maintenance unit, when setting the index information, is configured to: adding an index for identifying the media segment information in each segment of the TS code stream;

8. The system of claim 7, wherein the index information further comprises at least one of:

total length of RAP index information;

an RAP type for identifying a type of a corresponding segment RAP;

PID of the code stream carrying RAP.

9. The system according to any one of claims 6 to 8, wherein the data obtaining performing unit, when obtaining the frame at the RAP according to the media segment information included in the index information, is configured to:

10. The system according to claim 9, wherein the trick mode includes at least one of: