KR0164184B1 - Encoding controlling apparatus of a mpeg compression disc - Google Patents

Abstract

The present invention relates to an encoding control apparatus for moving picture compression disks. The video source 40 having an SIF image to be encoded, and a preprocessing unit 42 for performing preprocessing by filtering luminance and color difference signals of the SIF image. ), An adder 44 for adding the preprocessed video data and the frame data whose motion is compensated for, the original video data according to the I picture from the preprocessor 42 or the P picture through the adder 44. A switching unit 46 for selectively passing video data, a DCT unit 48 for performing DCT processing for converting video data passing through the switching unit 46 into coefficients in a frequency domain, and the DCT processed I picture; Filtering means (50, 52) for filtering the P value of the video data of the P picture into low frequency components (ILF, PLF) and high frequency components (IHF, PHF), and filtering the separated I pictures and P pictures. Separation switching unit 54 is switched so that the wave component (ILF, PLF) and the high frequency component (IHF, PHF) passes in sequence, the low frequency component (ILF, PLF) and the high frequency component of the separated I-picture and P-picture ( The quantizer 56 quantizes IHF and PHF, and estimates the motion of the quantized frame data through the video data of the preprocessor 42 to add frame data according to the P picture whose motion of the video screen is compensated. 44. The macroblock shape determining unit (58) includes motion estimation / compensation means (58) for transmitting the low frequency component (ILF, PLF) and high frequency component (IHF, PHF) of the quantized I-picture and P-picture. A variable length encoding unit 62 for DLC processing based on the motion vector from 60), a channel buffer 64 for buffering the DLC processed video stream data for each channel, and combining the video stream data and audio stream data. The program to be recorded on the video disc 68 Provided with a program stream generating unit 66 for generating a program data stream it is characterized in that configured.

Description

Encoding Control Device of Movie Compressed Disk

1 is a block diagram showing the structure of a conventional video disc encoding apparatus.

2 is a diagram illustrating a structure according to a picture group of video stream data included in a video disc.

3 is a diagram illustrating a state in which an I picture and a P picture of the picture group shown in FIG. 2 are separated and coded into a low frequency component and a high frequency component according to a preferred embodiment of the present invention.

4 is a diagram showing a state in which video stream data including an I picture and a P picture coded as shown in FIG. 3 are recorded on a video disc.

5 is a block diagram showing the configuration of an encoding control apparatus for a moving picture compressed disk according to the present invention.

* Explanation of symbols for main parts of the drawings

40: video source 42: preprocessor

44: adder 46: switching unit

48: DCT section 50: LPF

52: HPF 54: separation switch

56: quantization unit 58: motion estimation / compensation means

60: macroblock shape determining unit 62: variable length encoding unit

64: channel buffer 66: program stream generation unit

66: Video Disc

The present invention relates to an encoding control apparatus for moving picture compressed discs. More particularly, the present invention relates to an encoding control apparatus for encoding a moving picture compressed disc. The present invention relates to an encoding control apparatus for moving picture compressed discs, which enables a natural moving picture to be reproduced during decoding in the fast playback mode.

In general, an MPEG type information recording medium that records audio / video information using a moving picture expert group (MPEG) method, such as a video compact disc (hereinafter, referred to as a video disc), is attracting attention as a next generation audio / video information recording medium. As such, the MPEG-type information recording medium includes MPEG, that is, more precisely the standard specification for moving picture processing proposed by MPEG1 (when a video disc is about 74 minutes on the information recording surface of a 12 cm compact disc). Video and audio are recorded in a compressed state based on a read capacity of about 650 MB and a read speed of 1.5 Mbps per second).

In the MPEG method, while conventional JPEG removes spatial redundancy, since it is composed of moving images, VLC (Variable Length Coding) is removed by removing temporal redundancy through DCT (Discrete Cosine Transform) processing and quantization. Through this, the optimal compression rate is ensured and recorded on the video disc.

On the other hand, according to the encoding apparatus for encoding video stream data in such a manner that the video disc can be recorded on the video disc, as shown in FIG. 1, the SIF (Source Input Format) image to be encoded is encoded, that is, the luminance signal. A video source 2 in which a decimated image is generated with a unit of 720 × 480 and a coding unit of a color difference signal of 360 × 480 each having a number of 4: 2: 2, and the video source 2 Pre-processing unit that performs pre-processing to filter the horizontal and vertical axes of the SIF image from) to make the luminance signal 360 × 240 and the color difference signal 180 × 120. ) Is equipped.

Also, according to the same figure, an adder for adding SIF video input data according to an I picture preprocessed by the preprocessing unit 4 and video data according to P pictures whose motion is compensated from the motion compensator 22 to be described later. (6) and a switching section 8 which is switched to selectively pass the original video data according to the I picture through the preprocessor 4 or the video data according to the addition operation from the adder 6, DCT unit 10 for performing DCT (Discrete Cosine Transform) to remove spatial redundancy by converting spatial domain information of video data passing through switching unit 8 into coefficients consisting of frequency domain, and macroblocks obtained in DCT processing A quantization unit 12 for quantizing a coefficient value of an inverse, inverse quantization processing for using the quantized video data as reference frame data for subsequent frame data The inverse unit 14, an inverse DCT unit 16 for inverse-DCT converting the inverse quantized frame data, an adder 18 for adding inverse DCT-converted frame data and frame data whose motion is compensated for, and an adder 24 thereof. The first frame memory 20 which delays and stores the frame data added by the frame unit, and the estimated value from the motion estimation unit 10 which estimates the motion of the video screen by comparing the frame data of the video information with the preceding frame data. By using the motion compensation unit 22 to compensate for the movement of the frame data from the first frame memory 20 to generate frame data such as a P picture, the frame data estimated by the motion estimation unit 24 are obtained. Between the second frame memory 26, which is delayed stored in units of frames, the frame data from the first frame memory 20 and the frame data from the second frame memory 26, between 16 × 16 macroblocks.In other words, VLC (Variable Length Coding) is performed on the macroblock shape determining unit 28 and the quantization unit 12 to generate a motion vector, thereby forming video stream data, and various start bits. And a variable length encoder 30 for adding additional information, a channel buffer 32 for buffering the encoded encoded video stream data for each channel, and the program stream data by combining the video stream data with the audio stream data. And a program stream generation unit 34 for recording on the video disc 36.

In the encoding device of a video disc having such a component, the SIF image provided in the video source 2 is prefiltered by the preprocessor 4, and then the DCT unit 10 passes through the switching unit 8. DCT processing of the original video data converts the spatial domain information into coefficients of the frequency domain, and the quantization unit 12 quantizes coefficient values of the DCT-converted video data, while the quantized frame data is inverse quantized. After inverse quantization through 14 and inverse DCT conversion through the inverse DCT unit 16, the motion compensation unit 22 of the preceding frame data is estimated by the motion estimation unit 24 through an estimated value estimated by the motion estimation unit 24. It is added by the adder 18 together with the frame data which compensates for the movement, and is delayed and stored in the first frame memory 20.

In this state, the motion compensator 22 corrects the motion of the frame data from the first frame memory 20 based on the estimated value according to the motion estimation of the motion estimation unit 24, and thereby the frame data of the P picture. The adder 6 adds the original frame data according to the I picture from the preprocessor 4 and the frame data according to the P picture from the motion compensator 22 to add the frame data. The frame data passes through the switching section 8.

On the other hand, the frame data quantized by the quantizer 12 is VLC-processed by the variable length encoder 30 to be generated in the form of video stream data, and the video stream data buffered through the channel buffer 32 is programmed. The stream generator 34 is combined with the audio stream data to generate program stream data.

Accordingly, the video disc 36 has an I picture composed of the original reference picture, a P picture predicted based on the I picture or a previous P picture, and forward and backward and bidirectional prediction through the I picture or P picture. Video information data consisting of B pictures are inputted in the order of I picture-B picture-B picture-B picture-P picture-B picture-B picture-B picture-P picture ... and then I picture-P picture-B Encoded and recorded in the order of Picture-B Picture-B Picture-P Picture-B Picture-B Picture-B Picture ...

However, in the encoding device of such a video disc, video stream data having the reference I picture and the P picture predicted from the I picture and the B picture predicted from the I picture and the P picture is encoded and recorded on the video disc. Therefore, when the video disc is played at high speed by a special playback mode such as FF (Fast Forward) or FR (Fast Feverse), it is impossible to decode data before and after the P and B pictures. It is necessary to decode only the I picture, which means that the next I picture must be played after the screen playback of the I picture. Therefore, if the picture being played is unnaturally connected or the picture is severely moved, each screen is cut off. There are disadvantages.

Accordingly, the present invention has been made in view of the above circumstances, wherein an I picture of video stream data recorded on a video disc is separated into a high frequency component and a low frequency component, and the P picture is separated into a high frequency component and a low frequency component of the I picture. It is an object of the present invention to provide an encoding control apparatus for moving picture compressed discs, which enables video reproduction of a natural moving picture by decoding only low-frequency components of an I picture and a P picture during high-speed playback of the video disc.

In order to achieve the above object, according to the encoding control apparatus of a moving picture compression disk according to the present invention, a video source having an SIF image to be encoded, and a luminance signal and a color difference signal of the SIF image are filtered to perform preprocessing. A preprocessor, an adder for adding the preprocessed video data and frame data whose motion is compensated for, a switching unit for selectively passing the original video data according to the I picture from the preprocessor or the video data according to the P picture through the adder. And a DCT unit for performing DCT processing for converting the video data passing through the switching unit into coefficients in a frequency domain, and filtering them to be separated into a low frequency component and a high frequency component in coefficient values of the video data of the DCT-processed I-picture and P-picture. Filtering means, and the low frequency component of the filtered and separated I-picture and P-picture A separation switching unit which switches frequency components to pass sequentially, a quantization unit to quantize low frequency components and high frequency components of the separated I-picture and P-picture, and estimates the motion of the quantized frame data through video data of a pre-processing unit. Motion estimation / compensation means for sending frame data according to the P picture, which is compensated for the motion of the image screen, to the adder, and the frame data including the low frequency components and the high frequency components of the quantized I picture and the P picture to the macro block shape determiner. Variable length coding unit for DLC processing based on the motion vector from the channel, Channel buffer for buffering the DLC processed video stream data for each channel, Program stream data for combining the video stream data and audio stream data to be recorded on a video disc Video configured with a program stream generation unit for generating a It provides an encoding control device in the axial disc.

According to the present invention made as described above, the I picture component is separated into a low frequency component and a high frequency component when encoding video stream data recorded on a video disc, and the P picture component is P-coded high frequency component and I-coded low frequency. By separating and encoding the components, only the low frequency components of the I picture and the P picture are decoded during high-speed playback of the video disc so that the video can be reproduced at high speed as a moving picture.

Hereinafter, the present invention configured as described above will be described in detail with reference to the accompanying drawings.

That is, FIG. 3 is a diagram illustrating a state in which I pictures and P pictures of the picture group shown in FIG. 2 are separated and coded by a low frequency component and a high frequency component according to a preferred embodiment of the present invention. The I picture of the video stream data recorded on the video disc by the encoding operation is separated into low frequency components (ILF) and high frequency components (IHF), and the P pictures are encoded by the high frequency components (PLF) and the low frequency components (PHF). Therefore, I coded low frequency components are separated and coded.

FIG. 4 is a diagram showing a state in which video stream data including an I picture and a P picture coded as shown in FIG. 3 are recorded on a video disc, and according to the video stream data shown in FIG. The low frequency component (ILF) of the I picture is coded to be formed at the front of the high frequency component (IHF), and the low frequency component (PHF) of the P picture is coded to be formed at the front of the high frequency component (PHF). The B picture component is coded to be formed after the high frequency component (PHF), and the system header data including system information about the I picture and the P picture component is formed in front of the video stream data.

5 is a block diagram showing the configuration of an encoding control apparatus for moving picture compressed disks according to the present invention. In the present invention, reference numeral 40 denotes a decimation of a luminance signal and a color difference signal at 4: 2: 2. A video source with one SIF image is shown, and 42 denotes a preprocessor for filtering the SIF image and performing preprocessing for encoding.

Reference numeral 44 denotes an adder for generating a P picture by adding the video data from the preprocessor 42 to the motion compensated frame data from the motion estimation information means 58, which will be described later. The switching unit switches so that the raw video data corresponding to the I picture from the preprocessor 42 or the frame data corresponding to the P picture from the adder 44 are selectively passed.

Reference numeral 48 denotes a DCT unit for performing DCT processing to convert video data of an I picture or P picture through the switching unit 46 into coefficient values in a frequency domain, and 50 denotes a DCT processed I picture or P picture. The high frequency component is removed from the video data of the filter, and the low frequency component (ILF, PLF) is filtered to pass the LPF. 52 denotes the low frequency component from the video data of the DCT-treated I or P picture and the high frequency component (IHF, HPF which passes through only PHF).

Reference numeral 54 selectively selects a pass state of the low frequency components ILF and PLF of the I picture or P picture filtered by the LPF 50 and the high frequency components IHF and PHF filtered by the HPF 52. Switching so that the low frequency component (ILF) of the I picture is formed at the front of the high frequency component (IHF), and the low frequency component (PHF) of the P picture is formed at the front of the high frequency component (PHF). The separation switching unit to form a B picture component after the high frequency component (PHF) of.

Also, reference numeral 56 denotes a quantization unit for quantizing video data including high frequency components (IHF, PHF) and low frequency components (ILF, PLF) of the I picture and the P picture through the separation switching unit 54, and 58 Denotes a motion estimation / compensation means for generating frame data of a P picture whose motion of the screen is compensated on the basis of the motion estimation by inverse quantization and inverse DCT processing of the quantized video data.

Reference numeral 60 denotes a macroblock shape determination unit for determining the shape of the macroblock through the motion estimation value from the motion estimation / compensation means 58, and 62 denotes I quantized by the quantization unit 56. Represents a variable-length encoding unit for performing VLC processing on frame data according to high frequency components (IHF, PHF) and low frequency components (ILF, PLF) of pictures and P pictures based on the macroblock shape determining unit 60 to form video stream data. .

Reference numeral 64 denotes a channel buffer for buffering the video stream data for each channel, and 66 denotes program stream data for recording on the video disc 68 by combining the buffered video stream data with the audio stream data. Shows a program stream generation unit.

Next, the operation of the present invention made as described above will be described in detail with reference to FIG.

First, when the SIF image from the video source 40 is filtered and preprocessed by the preprocessor 42, the video data according to the I picture from the preprocessor 42 is switched. DCT processing through the DCT unit 48 through the) and the high frequency component (IHF) and low frequency component (ILF) is filtered and separated by the LPF (50) and HPF 52, respectively, separation switching unit 54 The coefficient value is quantized by the quantization unit 56 while passing in the order of the high frequency component IHF to the low frequency component ILF through the switching operation.

Then, the motion estimation / compensation means 58 inverse quantizes and inverse-DCT-processes the quantized frame data through the quantization unit 56, while the motion estimation / compensation means 58 applies the estimated values estimated by the previous frame data. As a result of compensating for the movement of the frame data, frame data according to the P picture is generated, and the adder 44 performs video data corresponding to the I picture from the preprocessor 42 and the motion estimation / compensation means 58. Frame data corresponding to the P picture is generated by adding the frame data compensated for by the motion from the circuit), while the frame data according to the P picture passes through the switching unit 46 and is applied by the DCT unit 48 to the DCT. Is processed.

At this time, the LPF 50 and the HPF 52 filter and separate the high frequency component (PHF) and low frequency component (PLF) of the DC-processed P picture by the DCT unit 48, and the P picture. The high frequency component PHF and the low frequency component PLF pass through the separation switching unit 54 in the order of the high frequency component PHF-low frequency component PLF and are quantized by the quantization unit 56.

Meanwhile, the variable length encoder 62 performs VLC processing on the high frequency components (IHF and PHF) and the low frequency components (ILF and PLF) of the quantized I and P pictures through the quantization unit 56 to form video stream data. The video stream data buffered through the channel buffer 64 is converted into program stream data combined with the audio stream data by the program stream generation unit 66 and recorded on the video disc 68.

According to the above-described present invention, the I picture and the P picture of the video stream data recorded on the video disc are separated from each other by the filtering operation into the low frequency component and the high frequency component, respectively, and are sequentially ordered from the low frequency component to the high frequency component. As encoding is performed, the video can be reproduced only by decoding the video by decoding only the low frequency components of the I picture and the P picture when decoding the video disc in the high speed playback mode. .

Claims (3)

A video source 40 having an SIF image to be encoded, a preprocessing unit 42 which performs preprocessing by filtering the luminance signal and the color difference signal of the SIF image, and the preprocessed video data and frame data whose motion is compensated for. An adder 44 which adds s, a switching unit 46 for selectively passing original video data according to the I picture from the preprocessor 42 or video data according to the P picture through the adder 44; A DCT unit 48 for performing DCT processing for converting video data passing through the switching unit 46 into coefficients in a frequency domain, and a low frequency component (ILF) in coefficient values for the video data of the DCT-processed I-picture and P-picture. Filtering means (50, 52) for separating into a (PLF) and a high frequency component (IHF, PHF), low-frequency components (ILF, PLF) and high-frequency components (IHF, PHF) of the filtered and separated I pictures and P pictures To pass through this sequence A separation switch unit 54 to be positioned, a quantization unit 56 for quantizing low frequency components (ILF, PLF) and high frequency components (IHF, PHF) of the separated I-picture and P-picture, and a pre-processing unit for the quantized frame data. Motion estimation / compensation means 58 for estimating the motion through the video data of 42 and transmitting frame data according to the P picture whose motion of the image screen is compensated to the adder 44, the quantized I picture and P A variable length encoder 62 for DLC processing the frame data including the low frequency component (ILF, PLF) and the high frequency component (IHF, PHF) of the picture based on the motion vector from the macroblock shape determiner 60; A channel buffer 64 for buffering the DLC processed video stream data for each channel, and a program stream generation unit 66 for generating program stream data for combining the video stream data and the audio stream data to be recorded on the video disc 68. With) Encoding control apparatus for a moving picture compression disc, characterized in that W is configured. The low frequency component (ILF, PLF) of the I picture and the P picture filtered by the filtering means (50, 52) is filtered by a low pass filter (LPF), and the high frequency of the I picture and the P picture. Component (IHF, PHF) encoding control device for a moving picture compressed disc, characterized in that the filter to be filtered by a high pass filter (HPF). The low frequency component (ILF, PLF) and the high frequency component (IHF, PHF) of the I picture and the P picture through the filtering means (50, 52) are ILF-IHF. -PLF-Encoding control device for moving picture compressed disks, characterized in that the switch is switched to pass in the order of PHF.