KR20020088633A - Transmission unit and Coding method for moving picture - Google Patents

Abstract

PURPOSE: An apparatus for transmitting moving pictures and a method for encoding moving pictures are provided to improve decoding capacity of encoded moving picture data, thereby increasing the reproduction probability of the moving picture data, reducing delay probability in transmission, realizing moving picture transmission through the Internet at low cost. CONSTITUTION: A method for encoding moving pictures includes the steps of forming a group of images by sorting moving pictures input from a moving picture input device by predetermined frame(S100), encoding a first image frame of each group image into intra images(S110), encoding image frames after the first image frame into inter images by calculating the difference with the first intra image of the corresponding group, which is a reference image(S120).

Description

Video transmission device and coding method {Transmission unit and Coding method for moving picture}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video transmission apparatus and an encoding method, and to a video encoding method for encoding a video input from a video input apparatus and a video transmission apparatus for transmitting an encoded digital video signal to a network.

Conventional video transmission apparatuses can be broadly classified into a web camera server apparatus and an Internet broadcasting system apparatus which transmit JPEG still images at regular time intervals to be viewed through a web browser of a client computer.

1 is a conceptual diagram of a server apparatus of a conventional web camera.

The server device of the web camera is a client that the web camera server 100 compresses the moving image data input through the web camera 200 into a JPEG file which is a still image compression standard and accesses the web camera server 100 via a network. Send to computer 300. The client computer 300 decodes and restores a JPEG file transmitted from the server 100 of the web camera using a JPEG decoder built in a web browser.

Specifically, when accessing the web server 101 embedded in the web camera server 100 by the HTTP transmission protocol through the web browser 301 of the client computer 300, the web server 101 is a predetermined web page Is transmitted to the web browser 301 of the client computer 300. This web page is written in HTML format, and assuming that a JPEG file is linked to the web page, the web server 101 is configured when the client clicks on the link portion of the JPEG file on the web browser 301. The JPEG file is transferred to the web browser 301 of the client computer 300, and the web browser 301 drives the built-in JPEG decoder to decode the transmitted JPEG file to play back the image. This is the same as transmitting a still image in general.

If you update the still image very quickly, it becomes a movie. That is, the web camera server 100 has a built-in JPEG encoder (102) to encode a video signal continuously input by the web camera 200, and then periodically (the very fast speed) of the JPEG file Update). The web server 101 continuously retransmits the updated JPEG file to the client computer 300, and the client computer 300 decodes and reproduces the JPEG file retransmitted through the JPEC decoder built in the web browser 301. Video is played.

The video transmission technology by the server apparatus of the web camera uses a transmission technology for still images in the video, and as shown in FIG. 2, each of the input images A1, A2, A3, A4, ..., An When encoded through the JPEG encoder of the camera server and output in the order of encoded images B1, B2, B3, B4, ..., Bn, the client computer 1 is connected to a wide bandwidth so that the encoded images B1, B2, B3, B4 After receiving all Bn, ..., Bn and decoding through JPEG decoder in web browser, output A1 ', A2', A3 ', A4', ..., An ', and client computer 2 is lower than client computer 1. It is connected to the bandwidth and receives only B1, B3, B5, and Bn-1 of the whole, and decodes it through a JPEG decoder in a web browser to output A1 ', A3', A5 ', ..., An-1'.

However, among A1, A2, A3, A4 ..., An which are close to each other, for example, adjacent A2, A3, A4 images are encoded as still images B2, B3, B4 even though the differences are very fine images. That is, there is a problem in that the amount of data generated by encoding for all still pictures is large.

In addition, the video transmission technology by the server device of the web camera is significantly lowered the frame rate when transmitting the same image to a plurality of client computers.

In addition, the transmission technology by the server device of the web camera transmits a JPEG file according to the HTTP transmission protocol, the HTTP transmission protocol is based on the TCP / IP method of the Internet protocol, the TCP / IP method is a transmission error As a protocol that guarantees that there is no error, as shown in FIG. 3, when the data to be transmitted is divided into packets, which are a predetermined unit, one packet is transmitted, and a signal that the packet has been successfully received from the receiving computer is received. If the next packet is transmitted, otherwise the packet is retransmitted. However, this transmission method not only slows down, but when multiple client computers connect, only one of the client computers needs to retransmit the packet if a reception failure occurs, causing other client computers to receive a delay. There was a problem.

In contrast, an internet broadcasting system apparatus encodes a video using a video encoding method such as MPEG, and then transmits the video to a client computer. The client computer temporarily stores the transmitted data in a memory and then executes an ActivX or Plug-in type decoding program. Drive and decode and play the video through the video player.

4 is a conceptual diagram of a conventional internet broadcasting system employing a motion coding method.

The internet broadcasting system includes an encoding and streaming server 100a for encoding an image signal input from a web camera 200, and receives encoded video data from the encoding and streaming server 100a to a plurality of client computers 300. The media server 100b for relaying and the web server 100c for transmitting the web page to the client computer 300 are separately installed.

In detail, when the web server 101 is connected to the web server 101 through an HTTP transmission protocol through the web browser 301 of the client computer 300, the web server 101 displays a predetermined web page of the client computer 300. Send to the web browser 301. This web page is written in HTML format, and assuming that a video file is linked to the web page, the client computer 300 when the client clicks on the link portion of the video file on the web browser 301 with a mouse. Receives the encoded video file from the media server 100b, decodes it, and plays it through a predetermined decoder program.

The encoding and streaming server 100a encodes a video signal input from the web camera 200, and may transmit the encoded video data directly to the client computer 300 by a simple streaming server therein. In order to connect with the computer 300, it is preferable to relay through the media server 100b which relays the encoded video transmission.

At this time, the media server 100b transmits a video file encoded by a streaming method according to the RTP transmission protocol, which is based on the UDP / IP transmission method, and the UDP / IP method is illustrated in FIG. 5. Unlike the TCP / IP method, the packet is transmitted unconditionally without checking whether the reception is successful from the receiving computer. Therefore, no matter how many client computers 300 are connected, retransmission due to reception failure does not occur.

Meanwhile, the encoded video data transmitted to the client computer 300 is stored in a buffer inside the client computer 300, and a predetermined amount is collected and then decoded. This is called buffering. Underflow or overflow should not occur in the buffer in order for the transmitted video data to be reproduced correctly. The state of this buffer is fed back to the media server 100b by a separate RTCP transmission protocol.

A feature of the motion encoding method such as MPEG is to classify the input video into several groups and classify the first video of each group, and to perform intra video encoding on the first video of each group, and to encode inter video by using the previous or subsequent video as a reference video. To do.

The intra picture encoding is almost similar to the still picture encoding. The image is divided into 16 × 16 pixel macroblocks, each macroblock is divided into 4 8 × 8 blocks, and then DCT (Descrete Cosine Trasform) conversion is performed. Quantization is performed, and the result is Variable Length Coding (VLC) coding using the Huffman technique. The encoded intra picture is called an I picture.

The inter-image encoding is a method of encoding only a component having a change by comparing each image with a previous or subsequent image. In this case, the case where only the previous image is compared is called a P-picture (Predictive Picture), and the case where a comparison between the previous image and the subsequent image is called a B-image (Bidirectional Predictive Picture). Reference Picture). In the case of such an inter image, the reference image must be reconstructed first for complete reconstruction. That is, when the reference image cannot be restored, the inter image is also not restored.

As described above, the techniques related to the Descrete Cosine Trasform (DCT) transformation, quantization, Huffman technique, and Variable Length Coding (VLC) coding are well known in the field of image processing and are variously described. Is omitted.

FIG. 6 shows an input image of an input image, which is grouped by an encoder into a group image, and then the first image and the last image of each group are intra-image coded into an I image, and the rest are compared to the previous image, respectively, and the inter image is a P image. Encoding is shown.

That is, I: 0 image is the first I image of each group, I: n + 1 is the last I image of each group, P: 1 image is encoded by I: 0 image as reference image, and P: n image is The P: n-1 image is encoded as a reference image.

In order to decode the encoded image, only those images in which the link is maintained are possible. In other words, intra I video can be decoded once received, but I: 0 video must be decoded to decode P: 1 video, and P: n-1 video must be decoded to decode P: n video. do. Therefore, the connection of the received image is disconnected from the client computer connected with the bandwidth that is insufficient to receive all such data, so that the P images after the disconnection are not properly restored.

For example, as shown in FIG. 7, if the client computer 1 is connected with a wide bandwidth and receives all video data from I: 0 to P: n, all received data is decoded and reproduced. However, if client computer 2 is connected with less bandwidth than client computer 1 and receives I: 0, P: 1, P: 3, ..., P: n-1, I: 0, P: 1 video Is correctly decoded and reproduced, but since there is no P: 2 image, subsequent P: 3, ... P: n-1 images cannot be decoded and cannot be reproduced.

That is, the Internet broadcasting system apparatus using the video encoding method has a much higher frame rate than the server apparatus of the web camera using the JPEG file as a still image, but the network having a bandwidth smaller than the amount of data generated by the encoding apparatus. There was a problem in that the client computer connected to the server could not deliver a proper video. The reason for this is that the encoding is performed on the premise of a specific bandwidth or more, and if it is smaller than this bandwidth, the client computer cannot receive all the encoded data, so that correct decoding cannot be performed.

In order to solve this problem, MPEG2 or MPEG4 introduces a technique called scalability so that a plurality of target bandwidths are used, and each client computer connects to the server with a bandwidth suitable for itself and receives and decodes a video.

The scalability is to enable the transmission of one input video to a plurality of client computers having a plurality of bandwidths, and to output an image quality or a frame rate corresponding to each bandwidth. In other words, better image quality or a larger number of image frames are output to a wideband client computer, and a smaller amount of image data is transmitted to a narrow bandwidth client computer through image quality deterioration or frame rate reduction.

The scalability supported by MPEG2 or MPEG4 includes spatial scalability, temporal scalability, and SNR scalability. Can be.

Of these, temporal scalability similar to the present invention will be described with reference to FIG. 8.

As shown, a sequence of input images is alternately divided into a basement layer and an enhancement layer. As described above, the images corresponding to the basement layer are grouped into a group image and separated into an intra image and an inter image to perform encoding. In other words, the basement layer itself forms a complete structure. The images of the enhancement layer are encoded using the images of the basement layer as reference images. Therefore, the image of the enhancement layer can be completely decoded only when combined with the images of the basement layer.

Considering this for each bandwidth, as shown in FIG. 9, the client computer 1 connected to the broadband network simultaneously receives and decrypts the basement layer and the enhancement layer, and the client computer 2 connected to the network having the low bandwidth performs the basement. By decoding and playing back only layers, each video data can be received and decoded and played according to bandwidth. In other words, when only the basement layer is received, a small number of video frames are played by the client computer. When both the basement layer and the enhancement layer are received, a large number of video frames are played. Adjusted.

This temporal scalability is useful when the bandwidth between the server and the client computer is known in advance. In other words, the client computer needs to determine the data to be transmitted in advance according to its bandwidth. However, this method does not work when the bandwidth between the server and the client cannot be determined, such as the Internet. That is, even when the basement layer only or the basement layer and the enhancement layer are set to receive both, connection may not be maintained at such a bandwidth in some cases. If you do not lose.

Therefore, the scalability technique assumes that a specific bandwidth is maintained, and it is not guaranteed to operate in a bandwidth-changing environment such as the Internet, and the encoder must perform encoding according to a plurality of bands from one input signal. The server had the problem of requiring more processing power.

The present invention has been invented to solve the above problems, and provides a video transmission apparatus and an encoding method capable of improving the decoding capability of video data transmitted to a plurality of client computers having an unspecified and variable network bandwidth. For that purpose.

It is still another object of the present invention to process a plurality of videos transmitted from two or more video input devices and transmit them to a plurality of client computers having unspecified and variable network bandwidths, thereby improving video decoding capability when decoding video data. It is to provide a transmission apparatus and an encoding method.

1 is a conceptual diagram of a server device of a conventional web camera

2 is a diagram of an encoding / decoding process of a server apparatus of a conventional web camera.

3 is a packet transmission process according to the TCP / IP method

4 is a conceptual diagram of a conventional internet broadcasting system employing a motion coding method.

5 is a packet transmission process according to the UDP / IP method

6 is a diagram illustrating a coding process in a conventional motion coding scheme.

7 is a diagram of decoding processing in a conventional motion coding scheme.

8 is a diagram illustrating encoding according to a conventional temporal scalability technique.

9 is a decoding process diagram according to a conventional temporal scalability technique

10 is a flowchart of a video encoding method according to the present invention.

11 is a decoding process diagram according to the present invention.

12 is a schematic diagram of a video transmission apparatus according to the present invention;

13 is an example of a video broadcasting system in a distributed environment in which a plurality of video transmission apparatuses according to the present invention are applied.

<Description of the symbols for the main parts of the drawings>

10: video transmission device 11: video signal input unit

12: encoder 13: network interface unit

14: control unit 20: video input device

30, 300: client computer 100: web camera server

101: web server 102: JPEG encoder

301: web browser

According to an aspect of a video encoding method according to the present invention for achieving the above object, the video encoding method according to the present invention is a video encoding method for encoding a video input from a video input device, input from the video input device The first video frame of each group image is encoded as an intra image, and the second and subsequent image frames of the group are separated from the first intra image of the corresponding group as the reference image. The calculation is characterized in that the encoding of the inter-image.

According to an aspect of a video transmission apparatus according to the present invention, a video transmission apparatus according to the present invention includes a video signal input unit for converting an analog video signal input from a video input device into a digital video signal; An encoder which compresses and encodes a digital video signal received from the video signal input unit; A network interface unit configured to transmit a compressed and encoded video signal received from the encoder to a network; A moving picture transmission device including a moving picture signal input unit, an encoding unit, and a control unit for controlling an overall device including a network interface unit, wherein the encoding unit groups digital video signals applied from the moving picture signal input unit in predetermined frame units. The first image frame of each group image is encoded into an intra image, and the second and subsequent image frames of each group image are encoded into an inter image by calculating a difference from the first intra image of the corresponding group as a reference image. It is characterized by processing.

According to an additional aspect of the video transmission apparatus according to the present invention, the signal input unit and the encoding unit of the video transmission apparatus according to the invention at least two or more signal input unit to process a plurality of videos transmitted from two or more video input device and It is characterized in that the encoder pair.

According to an additional aspect of the video transmission apparatus according to the present invention, all the intra video in each group encoded by the encoder is transmitted to a client computer to which the control unit of the video transmission apparatus according to the present invention connects with different bandwidths, respectively. The inter-images may be controlled to be output by adjusting a transmission frame rate according to the bandwidth of each of the client computers.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily understand and reproduce the present invention.

10 is a flowchart of a video encoding method according to the present invention.

First, in a group image forming step (S100), a group image is formed by grouping a video input from the video input device in predetermined frame units.

In operation S110, the first image frame of each group image is encoded as an intra image.

Finally, in the inter image encoding step (S120), the second and subsequent image frames of each group are encoded as inter images by calculating a difference from the first intra image of the corresponding group as the reference image.

Therefore, when the video encoded by the above method is transmitted from the server to the client computer through the network, the intra video can be restored regardless of whether the remaining video is received when the intra video is transmitted to the client computer. That is, referring to FIG. 11, the client computer 1 receives the entire group image, and the client computer 2 receives the intra image I: 0, I: n + 1 and the inter image P: 1, P: 3, ..., P Assuming that: n-1 is received, the client computer 1 recovers and reproduces all the images, and the client computer 2 can also recover all the transmitted images.

The reason for this is that unlike the conventional inter-image encoding before or after the reference image in the present invention, the reference image to the intra image I: 0 to each of the P: 1, P: 3, P: n-1 Since the inter picture is encoded, all inter pictures transmitted are decoded as long as the intra picture I: 0 is transmitted. Therefore, even if the image frame rate is output differently for each band, each client computer can efficiently decode and play encoded video data corresponding to its bandwidth.

The important thing here is that all the images in a group are very small in comparison, so that even if the intra image is encoded as the reference image as described above, the image is reproduced without being awkward when it is decoded and reproduced.

12 is a schematic diagram of a video transmission apparatus according to the present invention.

As illustrated, the video transmission apparatus 10 according to the present invention includes a video signal input unit 11, an encoding unit 12, a network interface unit 13, and a controller 14.

The video signal input unit 11 converts an analog video signal input from the video input device 20 into a digital video signal.

The encoder 12 compresses and encodes a digital video signal received from the video signal input unit 11.

At this time, the encoder 12 processes the digital video signal applied from the video signal input unit 11 to form a group image by uniting a predetermined frame unit, and the first image frame of each group image is encoded into an intra image. The second and subsequent image frames of each group are encoded as inter images by calculating a difference from the first intra image of the corresponding group as the reference image.

Since the intra image encoding technique and the inter image encoding technique have been described above, a detailed description thereof will be omitted.

The network interface unit 13 transmits the compressed and encoded video signal received from the encoder 12 to the network.

The controller 14 controls the apparatus including the video signal input unit 11, the encoder 12, and the network interface 13.

On the other hand, the controller 14 transmits all intra images in each group encoded by the encoder 12 to the client computer 30 connected with different bandwidths, and the inter images are transmitted to the client computer 30. The transmission frame rate is adjusted to fit the respective bandwidths so as to be output through the network interface unit 13.

Accordingly, the analog video signal input from the video input device 20 such as a web camera is converted into a digital video signal through the video signal input unit 11, and the digital video signal received from the video signal input unit 11 is converted into the digital video signal. The encoder 12 encodes according to the video encoding method described above.

The client is connected to the web server through a web browser of his client computer 30 in accordance with the HTTP transmission protocol to receive the web page, and when the mouse clicks the link portion of the video on the web page, the control unit 14 Under the control of the video transmission apparatus 10 outputs the video signal encoded by the encoder 12 through the network interface 13 to the client computer 30 in accordance with the RTP transmission protocol.

In this case, all intra images in each group encoded by the encoder 12 are transmitted to the client computer 30 connecting the video transmission apparatus 10 with different bandwidths, and the inter images are transmitted to the client computer 30. (30) When the transmission frame rate is adjusted and output to fit each bandwidth, the client computer 30 decodes and reproduces the same through a built-in video playback program.

In addition, according to an additional aspect of the present invention, the signal input unit 11 and the encoder 12 have at least two signal input unit and encoder pairs to process a plurality of moving images transmitted from two or more video input apparatuses. The cost per channel can be reduced by allowing a single device to handle multiple channels.

13 illustrates an example of a video broadcasting system in a distributed environment in which a plurality of video transmission apparatuses according to the present invention are applied.

This is to install the video transmission device according to the present invention in places scattered throughout the country, and to install a directory server, a web server, a media server in the central IDC (Internet Data Center), the video transmitted from the video transmission device of the scattered places throughout the country It is possible to build a broadcasting system having a plurality of video generation sources by allowing them to be accessed through a network through a client computer.

Therefore, by doing the above, the video transmission apparatus and the encoding method according to the present invention can achieve the above object.

As described above, the video transmission apparatus and the encoding method according to the present invention can improve the decoding capability when decoding the encoded video data transmitted to a plurality of client computers having an unspecified and variable network bandwidth, thereby increasing the reproduction probability of the video data. This high, low probability of transmission delay, and has a useful effect to implement the video transmission over the Internet at a low cost.

Although the present invention has been described with reference to the accompanying drawings, it will be apparent to those skilled in the art that various modifications may be made therein without departing from the scope of the invention, which is covered by the following claims.

Claims (4)

In the video encoding method for encoding a video input from the video input device, The video input from the video input device is bundled into a predetermined frame unit to form a group image, the first image frame of each group image is encoded as an intra image, and the second and subsequent image frames of each group are the first image of the corresponding group as the reference image. The video encoding method of claim 1, wherein the difference between the intra video and the intra video is calculated. A video signal input unit for converting an analog video signal input from a video input device into a digital video signal; An encoder which compresses and encodes a digital video signal received from the video signal input unit; A network interface unit configured to transmit a compressed and encoded video signal received from the encoder to a network; In the video transmission apparatus comprising a control unit for controlling the overall device including the video signal input unit, the encoder and the network interface unit, The encoding unit: The digital video signal applied from the video signal input unit is processed to form a group image by uniting a predetermined frame unit, the first image frame of each group image is encoded into an intra image, and the second and subsequent image frames of each group are reference images. And calculating the difference from the first intra picture of the corresponding group and encoding the inter picture. The method of claim 2, The signal input unit and the encoding unit: And at least two signal input and encoder pairs for processing a plurality of moving images transmitted from two or more moving image input apparatuses. The method of claim 2 or 3, The control unit: Each intra image in each group encoded by the encoder is transmitted to a client computer connected to each other with a different bandwidth, and the inter images are output by adjusting a transmission frame rate according to the bandwidth of the client computer. Video transmission apparatus characterized in that to control to.