JP2003299022A - Image data storing and reproducing system and method for storing and reproducing image data - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To suitably store image data arriving via a plurality of channels. <P>SOLUTION: An image data storing system for capturing, multiplexing and storing the image data compressed based on an interframe prediction via a plurality of channels comprises: an I picture extracting units 3-1 to 3-4 respectively provided at a plurality of channels to extract the independent image data without referring to other images in the image data; image format information extracting units 2-1 to 2-4 for retrieving the image format information; a format storage unit 7 for storing the retrieved image format information; a multiplexing unit 5 for multiplexing the extracted image data by aligning, coupling and adding a start position marker code to the head of the coupled image data group; and a storage device 6 for storing the data multiplexed by the unit 5. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video data storage system which captures video data compressed on the basis of inter-frame prediction through a plurality of channels and multiplexes and stores the video data. The present invention relates to a video data reproducing system for reproducing a video data, a video data storing method performed in the video data storing system, and a video data reproducing method performed in the video data reproducing system.

[0002]

2. Description of the Related Art In recent years, with the increase in capacity of communication networks, high quality video data has been transmitted. The video data transmitted on the network is
Usually, the data is compressed based on inter-frame prediction, and the bit capacity tends to increase as the image quality increases. Therefore, a large storage device is required to store and store data with high image quality, and there is a problem that the cost required for installation and operation increases.

By the way, in a video data storage system for video monitoring, it is rare that a video having a movement similar to that of a television is required, and in many cases, recording of a quasi-moving image in which a video change occurs several times per second. I wish I could.

[0004]

In response to the above-mentioned request of the video data storage system, the video data sent is MPEG video data or the like, and independent video data called an I picture, It is video data composed of a P picture that refers to the previous video and a B picture that refers to the previous and next videos. Therefore, even if the video data is thinned out at random, it is impossible to reproduce the video unless the video data that is the source of the difference data exists.

Therefore, conventionally, all video data has been recorded, and after a certain period of time has been erased, it has been recorded again on a secondary recording device such as a magnetic tape. However, when erased, there is a problem that the image cannot be restored, and when moved to the secondary storage device, there is a problem that the retrieval and reproduction of the image are troublesome.

Further, Japanese Patent Application No. 11-224656 discloses that I picture, P picture and B picture are divided into blocks and stored in a storage medium. But,
It does not correspond to the proper accumulation and reproduction of the video data that arrives via a plurality of channels, and the proper accumulation and reproduction of the video data from some surveillance cameras is required.

The present invention has been made to solve the problems of the conventional video data storage system as described above, and its object is to appropriately store and reproduce video data coming through a plurality of channels. A data storage / reproduction system and a video data storage / reproduction method are provided.

[0008]

A video data storage system according to the present invention is a video data storage system for capturing video data compressed on the basis of inter-frame prediction through a plurality of channels and multiplexing and storing the video data. Extraction means provided for each of a plurality of channels to extract independent video data that does not refer to other video from the video data, extraction means for extracting the video format information provided in the plurality of channels, and the extraction means. Format storage means for storing the extracted video format information and the video data extracted by the extraction means are arranged and connected, and at the head of the connected video data group,
A multiplexing means for adding and multiplexing a start position marker code including time information at the time of encoding extracted by the extracting means; and a multiplexed data storage means for storing the data multiplexed by the multiplexing means. It is characterized by having.

The video data reproducing system according to the present invention is
A video data storage system that captures video data compressed based on inter-frame prediction through a plurality of channels, multiplexes and stores the data, and is provided for each of the plurality of channels and refers to another video from the video data. Extraction means for extracting independent video data, extraction means for extracting the video format information provided in the plurality of channels, format storage means for storing the video format information extracted by the extraction means, and the extraction means Multiplexing means for arranging and concatenating the video data extracted by the above, and adding the start position marker code including the time information at the time of encoding extracted by the extracting means to the head of the connected video data group for multiplexing. And a multiplexed data storage for storing the data multiplexed by this multiplexing means. Based on the video format information stored in the format storage means, the detection means being attached to the video data storage system, and detecting the start position marker code from the data stored in the multiplexed data storage means. A decoding means for decoding the data stored in the multiplexed data storage means to obtain a reproduced image signal from independent video data for each channel; and the decoding based on the start position marker code detected by the detecting means. And a distribution means for distributing the reproduced image signal obtained by the means for each channel.

A video data storage method according to the present invention is a video data storage method in a video data storage system that captures video data compressed based on inter-frame prediction through a plurality of channels, multiplexes and stores the video data. An extraction step of extracting independent video data that is provided for each of the plurality of channels and that does not refer to other video from the video data; a step of extracting video format information from the plurality of channels; and a step of extracting video format information from the video data. Format storage step for storing the video format information that has been stored, and the video data extracted by the extraction step are connected side by side, and the time information at the time of encoding extracted by the extraction step is added to the head of the connected video data group. Start position marker code including A multiplexing step of multiplexing, characterized by comprising a multiplexed data storing step of storing the multiplexed data by the multiplexing step.

A video data reproducing method according to the present invention is a video data storage system for capturing video data compressed on the basis of inter-frame prediction through a plurality of channels and multiplexing and storing the video data, wherein the plurality of channels are stored. Extracting means for extracting independent video data which does not refer to other video from the video data, extracting means for extracting the video format information provided in the plurality of channels, and the video extracted by the extracting means. A format storing means for storing format information and the video data extracted by the extracting means are connected side by side, and the start of the head of the connected video data group includes the time information at the time of encoding extracted by the extracting means. Multiplexing means for adding a position marker code and multiplexing, and this multiplexing means In a video data reproducing method used for reproducing video data in a video data storage system comprising multiplexed data storage means for storing duplicated data, a start position marker code is generated from the data stored in the multiplexed data storage means. And a reproduction image signal from independent video data of each channel by decoding the data stored in the multiplexed data storage means based on the video format information stored in the format storage means. And a distribution step of distributing the reproduced image signal obtained by the decoding step to the channels based on the start position marker code detected by the detection step.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of a video data storage system, a video data reproduction system, a video data storage method and a video data reproduction method according to the present invention will be described below with reference to the accompanying drawings. FIG. 1 shows the configuration of a video data storage system according to the present invention. In this embodiment, four MPEG (Moving Picture Experts Group) (not shown)
Four streams (input video data) coming from the encoder via the network are multiplexed into one stream of I pictures and stored. 4 channels C
The input video data arriving at H1 to CH4 are input to the corresponding reception buffers 1-1 to 1-4, respectively.

The video data that arrives in this embodiment is
MPEG is adopted as a compression method based on inter-frame prediction. FIG. 2 shows a data structure of one general MPEG sequence. In the data for one sequence, a sequence header SH is arranged at the beginning, and then a GOP (Group of Pictures) header is arranged. The sequence header SH includes a sequence start code indicating a start point, a video size, a parameter used for quantization, and the like. The GOP header includes time information at the time of encoding.

After the GOP header, picture data continues in the order of IBBPBBP ... Although this order is not specified in the MPEG standard, it shows a widely used pattern. The number of I-pictures is 1 out of 15, and the cycle is 2 / second.

The reception buffers 1-1 to 1-4 that receive the MPEG video data as described above have the functions of removing network jitter and the like, and of preventing data loss due to the delay of the subsequent processing. Receive buffer 1
The video data held in -1 to 1-4 are taken in by the video format information extracting units 2-1 to 2-4 which constitute the taking-out means.

Video format information extraction units 2-1 to 2-
Reference numeral 4 decodes the sequence header SH and the GOP header to extract information unique to the stream such as video size and parameters used for quantization, and the rest to the I picture extraction units 3-1 to 3-4. It is what is sent. here,
The same video format is adopted for the streams arriving at the four video format information extraction units 2-1 to 2-4, and the video format information extracted by the video format information extraction unit 2-4 is stored in the format storage means. It is stored in the constituent format storage unit 7.

The I picture extracting units 3-1 to 3-4 function as an extracting means for extracting an I picture which is independent video data which does not refer to other video from the video data. I
The picture extracting units 3-1 to 3-4 write the extracted I picture in the corresponding picture buffers 4-1 to 4-4.

A multiplexing section 5 is connected to the picture buffers 4-1 to 4-4, and the multiplexing section 5 is a GOP obtained by the video format information extracting section 2-4 from the control section 8.
A start position marker code including a time code (time information) contained in the head is obtained, I pictures obtained from the four channels that are added to the head are arranged in order, concatenated, and multiplexed to form a multiplexed data storage means. The data is stored in the storage device 6 to be configured. The start position marker code including the time code (time information) is used to identify the head of the multiplexed stream and can also be used to search by the time code.

The MPEG stream analysis processing performed by the video format information extraction units 2-1 to 2-4 and the I picture extraction units 3-1 to 3-4 in the above video data storage system is performed as shown in the flowchart of FIG. Since the storage and storage process performed by the multiplexing unit 5 is performed as shown in the flowchart of FIG. 3, these processes will be described.

The MPEG stream analysis process and the storage / storage process are processes based on different time bases. That is, the MPEG stream analysis processing is performed at the timing of the data that is sent, and the video data of the I picture is written in the picture buffers 4-1 to 4-4 by this processing. On the other hand, in the storage and storage process, the picture buffers 4-1 to 4-
The I picture is read out from the memory 4 and multiplexed and stored in the storage device 6. In this way, the picture buffer 4-1
4-4 are accessed, the picture buffers 4-1 to 4-4 have an adjusting function so that writing and reading are not performed at the same time.

The MPEG stream analysis process shown in FIG. 3 is performed for one channel (the video format information extraction unit 2
-4) shows the processing. When the MPEG stream analysis process is started, the sequence start code is detected (S1), and the video format information is extracted by the detection and stored in the format storage unit 7 (S2). The storage in step S2 is performed only by the video format information extraction unit 2-4 because the same format is adopted in the four streams as described above. Next, after the picture start code is detected (S3), the picture type is analyzed by the detection and the I picture is detected (S4). When an I picture is detected, this I picture is written in the picture buffer 4-4 (S5). After that, the process returns to step S3 to continue the processing.

When the storage and storage process starts, the process corresponding to the flowchart shown in FIG. 4 is performed. That is,
It is detected whether it is the data accumulation timing (S11).
This data storage timing is given from the control unit 8. At the data accumulation timing, the I picture is read from each of the picture buffers 4-1 to 4-4 (S1
2) It is detected that one I picture has been read from all of the picture buffers 4-1 to 4-4 (S1).
3). When the I pictures are prepared from all of the picture buffers 4-1 to 4-4, they are multiplexed and the storage device 6
(S14).

In the above-mentioned multiplex storage (S14),
A start position marker code including a time code (time information) is sent from the control unit 8 to the multiplexing unit 5, and this is arranged at the beginning of the data for one cycle as shown in FIG. 5 (a). The time code (time information) is included in the GOP head, and the video format information extraction unit 2-
4 and is taken in by the control unit 8. In the multiplexed data for one cycle, each input channel CH1 to CHn is next to the start position marker code.
(In this case, n = 4), each I picture is arranged and multiplexed. The format storage unit 7 stores video format information as shown in FIG.

Next, an embodiment of the video data reproducing system according to the present invention will be described. This video data reproduction system is attached to the video data storage system shown in FIG. 1 and is configured as shown in FIG. A start position detection unit 21 that is connected to the storage device 6 and that constitutes a detection unit that reads the multiplexed data as shown in FIG. 5A and detects the start position marker code is provided.

An MPEG picture decoder 22 is connected to the start position detector 21. The MPEG picture decoder 22 is initialized by the video format information sent from the format storage unit 7, and is set to decode the I picture in the multiplexed data as shown in FIG. To be done. That is, the video format information includes various kinds of information necessary for obtaining a video signal from the MPEG system, such as the number of pixels and the number of lines of an image signal before being encoded by MPEG, and therefore initialization is performed based on this. As a result, the MPEG picture decoder 22 decodes the data stored in the storage device 6 based on the video format information stored in the format storage unit 7, and converts the independent video data for each channel into, for example, an NTSC signal or the like. And a decoding means for obtaining the reproduced image signal of.

An image signal distributor 23, which constitutes a distributor, is arranged at the subsequent stage of the MPEG picture decoder 22. At the output side of the image signal distributor 23, four channel image buffers 24-1 to 24-4 are provided. Are connected. The image signal distribution unit 23, based on the start position marker code supplied from the start position detection unit 21, reproduces the reproduced image signal for each channel in the image buffers 24-1 to 24-24.
Distribute to -4. That is, when the detection of the start position marker code is instructed by the start position detection unit 21, the distribution destination is the image buffer 24-1, and thereafter, the distribution destination is determined every time the detection of the picture start code is instructed by the MPEG picture decoder 22. Image buffer 24-2,
Switch to 24-3 and 24-4.

Through the above processing, the image buffer 24-
Video signals of channels CH1 to CH4 are stored in 1 to 24-4. Image buffer 24-1 to
The output of 24-4 is sent to the display device 25 and displayed as an image on the display. Display format instruction information is sent to the display device 25 from a personal computer or the like. The display device 25 uses the image signals stored in the image buffers 24-1 to 24-4 to perform the four-screen multi-display shown in FIG. 7A. 7B has a function of performing full-screen display on one screen shown in FIG. 7B by using any one of the stored image signals, and the display style shown in FIG. A display process is executed by selecting the display style of (b).

With the above configuration, the video data compressed based on the inter-frame prediction, which has a tendency that the bit capacity tends to increase as the video quality increases, is divided into a plurality of channels. Even when receiving and accumulating, the data amount can be reduced and the data can be stored and saved by a small-sized storage device. In addition, even during playback, by using independent video data (I picture in the above example) that does not refer to other video, the video is required to be monitored several times per second for video monitoring. Playback is performed to the extent of a quasi-moving image that causes changes. In the above example, the video format information is the same for all channels,
When the video format information is different in each channel, it may be taken out by the taking-out means, stored and used.

[0029]

As described above, according to the present invention, independent video data that does not refer to other video is extracted from the video data for each of a plurality of channels, and the extracted video data are arranged and connected. Since the start position marker code including the time information at the time of encoding is added to the beginning of the concatenated video data group and multiplexed and stored, the video data arriving via multiple channels should be properly accumulated and played back. Is possible.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a video data storage system according to the present invention.

FIG. 2 is a diagram showing a data structure according to MPEG, which is one of the compression methods of video data stored in the video data storage system according to the present invention.

FIG. 3 is a flowchart for explaining MPEG stream analysis processing in the video data storage system according to the present invention.

FIG. 4 is a flowchart for explaining a storage and storage process in the video data storage system according to the present invention.

FIG. 5 is a diagram showing a format of stored and stored data in the video data storage system according to the present invention.

FIG. 6 is a configuration diagram of a video data reproducing system according to the present invention.

FIG. 7 is a diagram showing a display example on the display device of the video data reproducing system according to the present invention.

[Explanation of symbols]

1-1 to 1-4 receive buffer 2-1 to 2-4 Video format information extraction unit 3-1 to 3-4 I picture extraction unit 4-1 to 4-4 picture buffer 5 Multiplexing unit 6 storage devices 7 format storage 8 control unit 21 Start position detector 22 MPEG picture decoder 23 Distributor 24-1 to 24-4 image buffer 25 display device

Claims (4)

[Claims] 1. A video data storage system in which video data compressed on the basis of inter-frame prediction is taken in via a plurality of channels, multiplexed and stored, and the video data storage system is provided for each of the plurality of channels, Extraction means for extracting independent video data that does not refer to video; extraction means provided in the plurality of channels for extracting video format information; and format storage means for storing the video format information extracted by the extraction means, The video data extracted by the extraction means are arranged and connected side by side, and a start position marker code including time information at the time of encoding extracted by the extraction means is added to the head of the connected video data group and multiplexed. Multiplexing means, and multiple storage means for storing the data multiplexed by this multiplexing means Video data storage system characterized by comprising a data storage means. 2. A video data storage system for capturing video data compressed on the basis of inter-frame prediction via a plurality of channels and multiplexing and storing the video data, the video data storage system being provided for each of the plurality of channels. Extraction means for extracting independent video data that does not refer to other videos,
Extraction means provided in the plurality of channels for extracting the video format information, format storage means for storing the video format information extracted by the extraction means, and video data extracted by the extraction means are connected side by side. Multiplexing means for adding the start position marker code including the time information at the time of encoding extracted by the extracting means to the head of the connected video data group and the data multiplexed by this multiplexing means And a detection means for detecting a start position marker code from the data stored in the multiplexed data storage means, the detection means being stored in the format storage means. Based on the video format information that is stored Decoding means for decoding the data stored in to obtain a reproduced image signal from independent video data for each channel; and a reproduced image signal obtained by the decoding means based on the start position marker code detected by the detecting means. A video data reproducing system comprising: a distribution unit that distributes the data corresponding to the channels. 3. A video data storage method in a video data storage system for capturing video data compressed based on inter-frame prediction through a plurality of channels and multiplexing and storing the video data, the video data storage method being provided for each of the plurality of channels. An extraction step of extracting independent video data that does not refer to other video from the video data, a step of extracting video format information from the plurality of channels, and storing the video format information extracted by the extraction step. The format storage step and the video data extracted in the extraction step are connected side by side, and a start position marker code including the time information at the time of encoding extracted in the extraction step is added to the head of the connected video data group. And a multiplexing step of multiplexing, Video data storage method characterized in that the multiplexing step includes a multiplexed data storing step of storing the multiplexed data. 4. A video data storage system for capturing video data compressed on the basis of inter-frame prediction through a plurality of channels and multiplexing and storing the video data, the video data storage system being provided for each of the plurality of channels. Extraction means for extracting independent video data that does not refer to other videos,
Extraction means provided in the plurality of channels for extracting the video format information, format storage means for storing the video format information extracted by the extraction means, and video data extracted by the extraction means are connected side by side. Multiplexing means for adding the start position marker code including the time information at the time of encoding extracted by the extracting means to the head of the connected video data group and the data multiplexed by this multiplexing means A video data reproducing method used for video data reproduction in a video data storage system comprising a multiplexed data storage means for storing a start position marker code from the data stored in the multiplexed data storage means. And stored in the format storage means A decoding step of decoding the data stored in the multiplexed data storage means on the basis of image format information to obtain a reproduced image signal from independent video data for each channel; and a start position marker detected by the detection step. And a distribution step of distributing a reproduced image signal obtained by the decoding step according to a code corresponding to channels.