KR100813989B1 - Coding and decoding method and apparatus using plural scanning patterns - Google Patents

Abstract

Disclosed are a coding method, a decoding method, and an apparatus using a plurality of scan patterns.

According to the present invention, a method of encoding image data includes: (a) obtaining N × M data by first source encoding the image data; (b) scanning the N × M data using a predetermined scan pattern selected from a plurality of scan patterns in response to the obtained N × M data; And (c) second source encoding the scanned data. As a result, encoding and decoding can be performed more efficiently on image data having various characteristics. In particular, it is possible to more efficiently encode and decode video data having an interlaced scanning image frame structure.

Description

Coding and decoding method using a plurality of scanning patterns, and apparatus therefor {Coding and decoding method and apparatus using plural scanning patterns}

1 is a block diagram of a conventional encoding apparatus;

FIG. 2 is a detailed block diagram of the second source encoder 700 of FIG. 1;

3 is a block diagram of an apparatus for decoding data encoded by the encoding apparatus of FIG. 1;

4 is a detailed block diagram of the second source decoder 710 of FIG. 3.

5 is an example of a zigzag scan pattern;

6 is an example of image data having an interlaced scanning frame structure;

7 is a block diagram of an encoding apparatus according to a preferred embodiment of the present invention;

8 is a block diagram of an encoding apparatus according to another preferred embodiment of the present invention;

9 is a block diagram of a decoding apparatus according to a preferred embodiment of the present invention;

10 to 12 illustrate examples of the scan pattern selector 93 of FIG. 9;

13 to 15 show scan patterns according to a preferred embodiment of the present invention,

16 to 17 are scan patterns according to another preferred embodiment of the present invention,

18 to 21 are reference diagrams illustrating a division scheme for using a plurality of scan patterns according to another exemplary embodiment of the present invention.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to encoding and decoding of image data, and more particularly, to a method for encoding, a method for decoding, and an apparatus using a plurality of scan patterns.

A series of video phone-related video coding methods, leading to H.261, H.263, H.263 +, etc., have been confirmed as the ITU-T international standard, and after that, technologies such as H.263 ++ Revision and MPEG-4 have been added. It became. However, these international standards do not meet new requirements for end-to-end delay time, coding efficiency, etc., so that ITU-T and ISO / IEC are the next generation coding method for next-generation video telephony. International standardization work is in full swing.

While conventional methods such as H.261 and H.263 target only images obtained by progressive scanning, the next generation encoding method includes images obtained by interlaced scanning as well. However, if the encoding method used for the progressive scan image is used for the interlaced scan image as it is or if the encoding method used for the interlaced scan image is used for the progressive scan as it is, the encoding efficiency may be deteriorated. This is because the sequential scan image and the interlaced scan image may have different characteristics of image data to be encoded due to different generation methods of the images.

1 is a block diagram of a conventional encoding apparatus.

Referring to FIG. 1, the encoding apparatus includes an encoding controller 100, a first source encoder 200, and a second source encoder 700. In addition, the first source decoder 300, the frame memory unit 400, the motion compensator 500, and the motion predictor 600 are provided.

When the image data is input, the encoding control unit 100 determines whether to perform motion compensation on the input image, that is, the coding type according to the characteristics of the input image or a predetermined operation purpose desired by the user, and generates a corresponding control signal. Outputs to the switch S10. When performing the motion compensation, since the image data inputted before or after is required, the first switch S10 is closed. When not performing the motion compensation, the first switch S10 is closed because the image data inputted before or after is not needed. S10 is opened. When the first switch S10 is closed, the difference image data obtained from the input image and the previous image are input to the first source encoder 200, and when the first switch S10 is opened, the input image is transferred to the first source encoder 200. Is entered.

The first source encoder 200 quantizes transform coefficient values obtained by transforming the input image data according to a predetermined quantization step to obtain N × M data. An example of the transform used may include a discrete cosine transform (DCT).

Meanwhile, since the image data input and encoded by the first source encoder 200 may be used as reference data for motion compensation of the image data input after or before, the first source encoder by the source decoder 300. After inverse quantization and inverse transformation, which are inverse processes of 200, the memory unit 400 is stored in the memory unit 400. At this time, if the data output from the source decoder 300 is difference image data, the encoding controller 100 closes the second switch S20 so that the difference image data output from the first source decoder 300 is a motion compensation unit. It is added to the output of 500 and then stored in the frame memory 400.

Meanwhile, the motion estimation unit 600 compares the input image data with data stored in the memory unit 400, finds the data most similar to the currently input data, and then compares the input image data with the input image data. Outputs a moving vector. When the motion vector is given to the memory unit 400, the memory unit 400 outputs the corresponding data to the motion compensation unit 500, and the motion compensation unit 500 encodes the current data based on the received data. Compensation values corresponding to the image data are created and output.

The second source encoder 700 encodes and outputs quantized transform coefficients output from the first source encoder 200. The motion vector encoder 900 receives and encodes information about a motion vector output from the motion predictor 600 and outputs the encoded information. The coding information encoder 800 receives and encodes coding type information, quantization step information, and other information necessary for decoding, which are provided by the encoding controller 100, and outputs the encoded information. The muxing unit 1000 outputs a bitstream finally obtained by multiplexing the outputs of the second source encoder 700, the coding information encoder 800, and the motion vector encoder 900.

In general, such an encoding method of an encoding apparatus includes a macroblock unit obtained by dividing an input image data into a predetermined size.

FIG. 2 is a detailed block diagram of the second source encoder 700 of FIG. 1.

Referring to FIG. 2, the second source encoder 700 includes a scan unit 701 and a variable length encoder 702. The scan unit 701 receives the N × M data including the quantized transform coefficient values and scans them in a zigzag pattern as shown in FIG. 5, and the variable length encoder 702 variably encodes the scanned data.

3 is a block diagram of an apparatus for decoding data encoded by the encoding apparatus of FIG. 1.

Referring to FIG. 3, the decoding apparatus includes a demux 110, a second source decoder 710, and a first source decoder 210 for demuxing a bitstream. A coding type information analyzing unit 120 and a motion vector analyzing unit 130 for analyzing the coding type information are provided.

The bitstream is demuxed into quantized transform coefficients, motion vector information, coding type information, and the like entropy coded by the demux 110. The second source decoder 710 entropy decodes the encoded transform coefficients and outputs quantized transform coefficients. The first source decoder 210 source decodes the quantized transform coefficients. That is, the reverse process of the first source encoder 200 of FIG. 1 is performed. For example, if the first source encoder 200 performs the DCT, the first source decoder 210 performs an Inverse Discrete Cosine Transform (IDCT). Thus, the video data is restored. The reconstructed image data is stored in the memory unit 410 for motion compensation.

On the other hand, the coding type information analyzer 120 detects the coding type and closes the third switch S30 in the case of an inter type requiring motion compensation, and then the motion compensator 510 to the data output from the first source decoder 210. Reconstructed image data is obtained by adding the motion compensation value output from The motion vector analyzer 130 informs the position indicated by the motion vector obtained from the motion vector information, and the motion compensator 510 generates and outputs a motion compensation value from the reference image data indicated by the motion vector.

4 is a detailed block diagram of the second source decoder 710 of FIG. 3.

Referring to FIG. 4, the second source decoder 710 includes a variable length decoder 703 and an inverse scan unit 704, and the reverse process of the process performed by the second source encoder 700 of FIG. 2. Do this. The variable length decoder 703 variable length decodes the variable length coded quantized transform coefficients to restore the N × M data, and the inverse scan unit 704 uses the zigzag scan pattern as shown in FIG. Reverse scan.

However, the conventional encoding apparatus and the decoding apparatus may cause a problem when encoding or decoding the interlaced scan image. Since the interlaced scan image has a time difference between each field, the image content may be changed even between adjacent fields. This is especially true for moving images. In particular, when an interlaced scan image is encoded, instead of being encoded in a field unit but encoded in a frame unit, that is, when a picture structure that is a unit for encoding is an interlaced frame structure, objects present in the image are top field data and bottom field. Since the position changes according to the data, the boundary line is distorted, and the data characteristic is remarkably changed in the vertical direction.

6 is an example of image data having an interlaced scanning frame structure.

Referring to FIG. 6, data existing in odd-numbered rows of the first, third, etc. of the image data constituting the N × M block are image data from the top field of the interlaced scan image, and second, four Data present in the first and even rows are image data from a bottom field. If there is a time difference between the top field and the bottom field and the motion corresponding to the time difference is large, the same object is composed of an image in which the same object is not aligned in the vertical direction. When transform-coding such data, high frequency components increase in the vertical direction, and non-zero transform coding coefficients may frequently appear in the lower row of the N × M block composed of the transform coding coefficients. When scanning using the zigzag scan pattern of FIG. 5, the characteristic that a non-zero transform coefficient frequently occurs in the vertical direction is not sufficiently reflected, resulting in a problem in that encoding efficiency is lowered. The same is true for images with a lot of motion and images with a large change in the vertical direction.

As described above, the conventional encoding method performs a scan using a single scan pattern, and thus there is a problem in that the encoding efficiency of image data having various characteristics cannot be maximized.

Accordingly, an object of the present invention is to provide an encoding method, a decoding method, and an apparatus capable of performing encoding on image data having various characteristics more efficiently.

Another object of the present invention is to provide an encoding method, a decoding method, and an apparatus capable of more efficiently performing encoding on image data having an interlaced scanning image frame structure.

It is still another object of the present invention to provide an encoding method, a decoding method, and an apparatus capable of performing encoding on an image having a lot of motion and an image having a severe change in a vertical direction.

According to the present invention, there is provided a method of encoding video data, the method comprising: (a) obtaining N × M data by first source encoding the video data; (c) scanning the N × M data using a predetermined scan pattern selected from a plurality of scan patterns in response to the obtained N × M data; And (d) second source encoding the scanned data.

A method of encoding video data, the method comprising: (a) obtaining N × M data by first source encoding the video data; (b) dividing the obtained N × M data into a plurality of regions; (c) scanning the N × M data using a scan pattern selected from a plurality of scan patterns in response to the divided region; And (d) second source encoding the scanned data.

Step (a) may include (a1) transcoding and encoding the image data; And (a2) quantizing the converted data to obtain the N × M data.

The step (c) may include (c11) selecting a scan pattern to be scanned later in an area in which a large number of substantially zero component values are present among the N × M data; And (c2) scanning the N × M data with the selected scan pattern.

Step (b) includes (b1) dividing the N × M data in a horizontal direction to obtain at least two sub-areas, or (b2) dividing the N × M data in a vertical direction to at least two sub-areas. And obtaining (b3) the N × M data in the vertical direction and the horizontal direction, respectively, to obtain at least four sub-regions.

The method of decoding image data, the method comprising: (a) generating scan pattern selection information for selecting a predetermined scan pattern from among a plurality of scan patterns; (b) reverse scanning the variable length decoded image data into N × M data using at least one scan pattern selected based on the scan pattern selection information; And (c) decoding the N × M data.

Step (c) may include (c1) inverse quantization of the N × M data; And (c2) transform-decoding the dequantized N × M data.

A method of decoding image data, the method comprising: (a) variable length decoding the image data; (b) inversely scanning the variable length decoded data using a predetermined scan pattern selected from the plurality of scan patterns to obtain N × M data; And (c) decoding the N × M data.

The step (b) may include (b11) selecting a scan pattern to be scanned later in an area in which a large number of substantially zero component values are present among the N × M data; And (b2) backscanning using the selected scan pattern.

The step (b) may include (b21) selecting a scan pattern that scans a horizontal high frequency component value relatively earlier than a vertical high frequency component value when the N × M data is based on image data of an interlaced scanning frame structure; And (b2) backscanning with the selected scan pattern.

In the step (b), (b31) the scan pattern for scanning the vertical frequency component value and the horizontal frequency component value in substantially equal order when the N × M data is obtained from the image data of the interlaced field structure or the progressive scan frame structure. Selecting a; And (b2) backscanning with the selected scan pattern.

Step (b) may include (b31) selecting a scan pattern corresponding to the macroblock type of the N × M data; And (b2) backscanning with the selected scan pattern.

Step (b) may include (b41) selecting a scan pattern corresponding to the picture structure and macroblock type of the N × M data; And (b2) backscanning with the selected scan pattern.

The step (a) preferably includes (a1) converting a code word constituting the image data into predetermined symbol data.

A method of decoding image data, the method comprising: (a) obtaining symbol data by variable length decoding the image data; (b) obtaining transform coefficient values from the symbol data; (c) generating scan pattern selection information for selecting a predetermined scan pattern from among the plurality of scan patterns; (d) reverse scanning the transform coefficient values into N × M data using a scan pattern according to the generated scan pattern selection information; And (e) decrypting the backscanned data.

Step (c) includes (c1) receiving and analyzing picture structure information; And (c21) generating, as a result of the analysis, the scan pattern selection information for selecting a scan pattern that scans a horizontal high frequency component value relatively earlier than a vertical high frequency component value in the case of an interlaced scanning frame structure, or (c1) a picture. Receiving and analyzing structure information; And (c22) generating scan pattern selection information for selecting a scan pattern that scans the vertical frequency component values and the horizontal frequency component values in substantially equal order in the case of the interlaced field structure or the progressive scan frame structure. It is preferable to include.

According to another aspect of the present invention, there is provided an apparatus for encoding video data, the apparatus comprising: a first encoder which encodes the video data to obtain N × M data; An encoding control unit providing scan pattern selection information for selecting a predetermined scan pattern among a plurality of scan patterns in response to the N × M data; And a second encoder which scans the N × M data by using the scan pattern according to the scan pattern selection information, and encodes the scanned data.

The first encoder may include a transform encoder for transform encoding the image data; And a quantizer for quantizing the data converted by the transform encoder and outputting the N × M data.

In addition, the above object is an apparatus for decoding image data, the apparatus comprising: a second decoder to obtain transform coefficient values by decoding the image data; A scan pattern selection unit which generates scan pattern selection information for selecting a predetermined scan pattern from among a plurality of scan patterns; And an inverse scanning unit which inversely scans the transform coefficient values into N × M data using the scan pattern according to the generated scan pattern selection information.

According to another field of the present invention, the above object is also achieved by a computer readable information storage medium having recorded thereon program code for performing the encoding method or the decoding method.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

7 is a block diagram of an encoding apparatus according to a preferred embodiment of the present invention.

Referring to FIG. 7, the encoding apparatus is a device for second source encoding according to the present invention, and includes a scan unit 71 and a variable length encoder 72. Reference numeral 73 is a switching unit 73 for selecting at least one scan pattern among a plurality of scan patterns 1, ..., k, ..., K according to a scan pattern selection signal input from the outside. In this embodiment, the scan pattern selection signal is input from an encoding control unit (not shown). The scan pattern refers to reference information in which the order of reading the N × M data is determined.

The scan unit 71 scans the N × M data input from the outside using at least one scan pattern selected by the switching unit 73 based on the scan pattern selection signal. That is, the N × M data is read in a predetermined order and converted into symbol data. The variable length encoder 72 variable length encodes the scanned data, that is, the symbol data.

In the present embodiment, the N × M data refers to block data obtained by encoding first image data. In this case, the first source encoding refers to performing transformation (and quantization) of two-dimensional image data composed of a plurality of pixel values as data of another domain. For example, image data composed of pixel values is converted into data in the frequency domain by DCT conversion. By the first source encoding, the image data can be represented as data of smaller size while containing substantially the same information. In the present embodiment, the symbol data refers to run-level data obtained based on reading N × M data using a predetermined scan pattern. The run-level data may be one-dimensional data obtained by separating the run value and the level value, or two-dimensional data in which the run value and the level value are paired, and the three-dimensional data including the last value, the run value, and the level value. It may be data. The variable length encoder 72 encodes a code word corresponding to 2D run-level data (Run, Level) (2D variable length coding), or generates 1D run-bell data in which Run components and Level components are separated. The code is coded with the corresponding code word (1D variable length coding). 3D variable length coding is also possible, which is coded by code words corresponding to 3D run-level data (LAST, Run, Level). At this time, the variable length encoder 72 may encode using a Huffman encoder or an arithmetic encoder.

The encoding controller has logic for selecting a scan pattern, and provides the scan pattern selection information to the switching unit 73 accordingly. According to the scan pattern selection logic, the encoding control unit provides scan pattern selection information for selecting a scan pattern to be read later in an area where a large number of substantially zero component values among N × M data exist. In particular, when N × M data is obtained from image data having an interlaced scan frame structure, scan pattern selection information for selecting a scan pattern that reads a horizontal high frequency component value relatively earlier than a vertical high frequency component value relative to a zigzag pattern is provided. Provides scan pattern selection information for selecting scan patterns that read vertical frequency component values and horizontal frequency component values in substantially equal order when N × M data is obtained from image data of interlaced scan field structure or sequential scan frame structure. do.

Alternatively, the encoding controller may provide scan pattern selection information for selecting a scan pattern corresponding to the macroblock type of the N × M data, or select a scan pattern corresponding to the picture structure and the macroblock type of the N × M data. Scan pattern selection information may be provided. Alternatively, the encoding control unit may provide scan pattern selection information for selecting different scan patterns for at least two sub-areas obtained by dividing the N × M data in the horizontal direction, or obtain the N × M data by dividing it in the vertical direction. Provide scan pattern selection information for selecting different scan patterns for at least two sub-areas, or different scan patterns for at least four sub-areas obtained by dividing N × M data in the vertical and horizontal directions, respectively The scan pattern selection information for selecting may be provided. Alternatively, N × M data may be arbitrarily divided into at least two sub-regions, and then scan pattern selection information for selecting different scan patterns for each sub-region may be provided.

8 is a block diagram of an apparatus for encoding image data according to a preferred embodiment of the present invention. However, for the block having the same function as that of FIG. 7, the same reference numerals will be given, and repeated descriptions will be omitted.

Referring to FIG. 8, the encoding apparatus may include a first source encoder 2 that obtains N × M data by source encoding image data, and an encoding controller that provides scan pattern selection information indicating a selected scan pattern among a plurality of scan patterns. (1) and scanning the N × M data using a scan pattern selected to correspond to each sub area when the N × M data is divided into a plurality of sub areas according to the scan pattern selection information. A second source encoder 7 for entropy encoding data is provided. In addition, the first source decoder 3, the memory unit 4, the motion compensator 5, the motion predictor 6, the coding information encoder 8, the motion vector encoder 9, and the muxing unit (10).

The input image data is composed of blocks obtained by dividing frames or frames input from a camera along a time axis into a predetermined size. The frame includes a sequential scan frame obtained by the sequential scan method, a field obtained by the interlaced scan method, or an interlaced scan frame. Therefore, the video data described below means pictures of progressive scan frames, interlaced scan frames, fields, and block structures.

When image data is input, the encoding control unit 1 determines a coding-type (intra coding / inter coding) according to whether to perform motion compensation on the input image according to a characteristic of the input image or a predetermined operation purpose desired by the user. To output the corresponding control signal to the first switch S1. When performing the motion compensation, the first switch S1 is closed because the image data inputted before or after is required, and when the motion compensation is not performed, the first switch S1 is not needed. S1 is opened. When the first switch S1 is closed, the difference image data obtained from the input image and the before or after image are input to the first source encoder 2, and when the first switch S1 is opened, only the input image is the first source encoder ( 2) is entered.

On the other hand, the encoding control unit 1 outputs the scan pattern selection information to the second source encoding unit 7. The scan pattern selection information is the same as that described with reference to FIG. Further, the encoding control unit 1 transmits or stores the scan pattern selection information to the transmitting end, if necessary.

The first source encoder 2 quantizes transform coefficient values obtained by transform encoding the input image data according to a predetermined quantization step to obtain N × M data that is two-dimensional data composed of quantized transform coefficient values. . An example of the transform used may include a discrete cosine transform (DCT). Quantization is performed according to a predetermined quantization step.

Meanwhile, since the image data input and encoded by the first source encoder 2 may be used as reference data for motion compensation of the image data input after or previously, the first source by the first source decoder 3 After the inverse quantization and inverse transform encoding, which are inverse processes of the encoder 2, are stored in the memory 4. In addition, if the data output from the first source decoder 2 is difference image data, the encoding controller 1 closes the second switch S2 so that the difference image data output from the first source decoder 3 is motion compensated. It is added to the output of section 5 and then stored in the memory 4.

Meanwhile, the motion estimation unit 6 compares the input image data with the data stored in the memory 4 to find the most similar data to the currently input data, and then compares the input image data with the motion vector. Output MV (Motion Vector). The motion vector is obtained with reference to at least one picture. That is, it may be calculated with reference to a plurality of past and / or future pictures. When the motion vector is transferred to the memory 4, the memory 4 outputs corresponding data to the motion compensation unit 5, and the motion compensation unit 5 currently encodes the image based on the received data. Compensation values corresponding to data are created and output.

The second source encoder 7 receives the quantized transform coefficients output from the first source encoder 2 and converts the N × M data into scan patterns selected according to scan pattern selection information provided from the encoding controller 1. Scanning then entropy encoding. The motion vector encoder 9 receives and encodes information about a motion vector output from the motion predictor 6. The coding information encoder 8 receives entropy coding by receiving coding type information, quantization step information, and other information necessary for decoding. The muxing unit 10 receives the outputs of the coding information encoder 8, the second source encoder 7, and the motion vector encoder 9, respectively, and outputs the finally obtained bitstream.

9 is a block diagram of a decoding apparatus according to a preferred embodiment of the present invention.

Referring to FIG. 9, the decoding apparatus is a device for decoding a second source according to the present invention and includes an inverse scan unit 91, a variable length decoder 92, and a scan pattern selector 93. Reference numeral 94 is a switching unit that selects at least one scan pattern from among the plurality of scan patterns 1, ..., k, ..., K in accordance with the scan pattern selection information provided from the scan pattern selection unit 93.

The variable length decoder 92 variable length decodes the input bitstream and then converts the input bitstream into symbol data. The inverse scan unit 91 rescans the symbol data using at least one scan pattern selected by the switching unit 94 based on the scan pattern selection information and reconstructs the N × M data. Thus, N × M data which is the first source coded block data is obtained. In the present embodiment, symbol data refers to run-level data. The variable length decoder 92 decodes a predetermined code word included in the bitstream to obtain corresponding run-level data.

The scan pattern selection unit 93 includes scan pattern selection logic to generate scan pattern selection information by itself and provide it to the switching unit 94. Hereinafter, the features of the scan pattern will be described based on the operation in the encoding process in consideration of the convenience of description. According to the scan pattern selection information, a region in which there are substantially zero component values among the N × M data to be reconstructed selects a scan pattern to be read later. In particular, according to the scan pattern selection information, when the N × M data to be reconstructed is based on the image data of the interlaced scanning frame structure, a scan pattern that reads the horizontal high frequency component value relatively earlier than the vertical high frequency component value is selected compared to the zigzag pattern. When the N × M data to be reconstructed or reconstructed is based on image data of an interlaced scan field structure or a sequential scan frame structure, a scan pattern that reads vertical frequency component values and horizontal frequency component values in substantially equal order is selected.

In addition, the scan pattern selection unit 93 provides scan pattern selection information for selecting a scan pattern corresponding to the macro block type of the N × M data to be reconstructed, or provides a picture structure and macroblock type of the N × M data to be reconstructed. Scan pattern selection information for selecting a corresponding scan pattern may be provided. Alternatively, scan pattern selection information for selecting different scan patterns may be provided for at least two sub-areas obtained by dividing the N × M data to be reconstructed in the horizontal direction, or the N × M data to be reconstructed is divided into the vertical direction. Provide scan pattern selection information for selecting different scan patterns respectively for at least two sub-areas, or different for at least four sub-areas obtained by dividing the N × M data to be reconstructed respectively in the vertical and horizontal directions, respectively. Scan pattern selection information for selecting a scan pattern may be provided.

In particular, a picture indicating whether a coding unit of the image data input in the encoding process is a frame structure or a field structure, or whether the input image data is a sequential scan image or an interlaced scan image, that is, a picture indicating the picture structure of the input image data The macroblock type information indicating the structure information or the macroblock type is information that the encoding apparatus must transmit to the decoding apparatus for decoding. Therefore, if the scan pattern selection logic is determined in advance so that the predetermined scan pattern can be selected based on the information, the coding apparatus can select the scan pattern without sending information for selecting the scan pattern to the decoding apparatus.

10 to 12 are exemplary implementations of the scan pattern selector 93 of FIG. 9.

Referring to FIG. 10, the scan pattern selector 93 receives picture structure information, generates and outputs scan pattern selection information corresponding to the scan pattern selection logic, or receives macroblock type information, and scan patterns corresponding thereto. Generate and output selection information.

Referring to FIG. 11, the scan pattern selection unit 93 receives picture structure information and macro block type information, and generates and outputs scan pattern selection information corresponding thereto according to the scan pattern selection logic.

Referring to FIG. 12, the scan pattern selector 93 receives a plurality of pieces of encoding information 1, 2,..., N, and generates and outputs scan pattern selection information corresponding to the scan pattern selection logic. Here, the encoding information is condition information on encoding, and means information provided from the transmitting end to the decoding apparatus for decoding.

10 to 12, the scan pattern selection logic may be implemented as a pre-stored mapping table. In the mapping table, mapping information in which a predetermined scan pattern is mapped to predetermined encoding information may be recorded. For example, corresponding scan patterns for macro block type information and picture structure information may be mapped.

13 to 15 are scan patterns according to an embodiment of the present invention.

13 to 15, scan patterns usable by the encoding apparatus or the decoding apparatus described with reference to FIGS. 7 to 9 are illustrated. For example, the encoding control unit 1 of the encoding apparatus or the scan pattern selection unit 93 of the decoding apparatus may have a probability that the vertical high frequency component value is 0 when the picture structure is an interlaced scanning frame structure. Scan pattern 4 can be selected which reflects a lower than probability. In other cases, scan pattern 3 can be selected.

16 and 17 are scan patterns according to another embodiment of the present invention.

16 and 17, a scan pattern for simultaneously scanning a plurality of N × M blocks, that is, a plurality of N × M data as one block is shown. According to this, the component values of the plurality of N × M data 1, 2, 3, and 4 are treated as component values of one N × M data and scanned. The number and size of the N × M data regarded as one block may be variously set as necessary, and the scanning order accordingly may be adjusted accordingly.

18 through 21 illustrate a division scheme for using a plurality of scan patterns according to a preferred embodiment of the present invention.

Referring to FIGS. 18 through 21, a method of dividing N × M data by the encoding apparatus or the decoding apparatus described with reference to FIGS. 7 through 9 to use a plurality of scan patterns is illustrated. For example, as shown in FIG. 16, the encoding control unit 1 of the encoding apparatus or the scan pattern selection unit 93 of the decoding apparatus may include at least two sub-regions obtained by dividing N × M data in a horizontal direction. Scan pattern selection information for selecting different scan patterns for each other, or as shown in FIG. 17, for each of at least two sub-regions obtained by dividing the N × M data in the vertical direction, respectively, Scan pattern selection information for providing scan pattern selection information or selecting different scan patterns for at least four sub-regions obtained by dividing N × M data in the vertical direction and the horizontal direction, respectively, as shown in FIG. 18. Can be provided. Furthermore, as shown in FIG. 19, scan pattern selection information for selecting different scan patterns may be provided for at least two sub-regions in which N × M data is divided diagonally. Alternatively, N × M data may be arbitrarily divided into at least two sub-regions, and then scan pattern selection information for selecting different scan patterns for each sub-region may be provided. Alternatively, a plurality of N × M blocks may be regarded as one block, arbitrarily divided into regions, divided into at least two sub-regions, and scan pattern selection information for selecting different scan patterns for each sub-region may be provided. Can be.

On the other hand, unlike the above-described example, the scan pattern selection unit 93 of the decoding apparatus receives scan pattern selection information provided from the encoding apparatus or reads scan pattern selection information stored by the encoding apparatus and based on the scan pattern selection information stored therein. It can also be configured to output.

The above-described encoding method and decoding method can be applied to multidimensional data as it is.

Furthermore, the above-described encoding method and decoding method can be created by a computer program. Codes and code segments constituting the program can be easily inferred by a computer programmer in the art. In addition, the program is stored in a computer readable medium, and read and executed by a computer to implement an encoding method and a decoding method. The information storage medium includes a magnetic recording medium, an optical recording medium, and a carrier wave medium.

As described above, according to the present invention, encoding and decoding can be performed more efficiently on image data having various characteristics. In particular, it is possible to more efficiently encode and decode video data having an interlaced scanning video frame structure. Furthermore, the efficiency is also increased for images with a lot of motion and images with a large change in the vertical direction.

In the future, the main areas for which the present invention is expected to be used are real-time interactive applications of video telephony, audio / video communication in mobile networks, video application services on the Internet, video transmission for real-time sign language and lip-reading communication, and video on demand. Video storage and retrieval for services, video storage and transmission applications for videomail applications, multi-point communication over heterogeneous networks, digital broadcasting, etc., the scope of application of the present invention is very diverse.

Claims (21)

In a method of encoding video data, (a) first N-M encoding the image data to generate N × M data; (c) scanning the N × M data using a selected scan pattern among a plurality of scan patterns; And (d) second source encoding the scanned data, And the scan pattern is selected based on a picture format of the N × M data. The method of claim 1, Step (a) is (a1) converting the image data; And and (a2) quantizing the transformed image data to generate the N × M data. The method of claim 1, And the picture format is an interlaced scan field format or a sequential scan field format. In the method for decoding video data, (a) generating scan pattern selection information for selecting one of the plurality of scan patterns; (b) reverse scanning the variable length decoded image data into N × M data using at least one scan pattern selected based on the scan pattern selection information; And (c) source decoding the N × M data, And the scan pattern selection information is generated based on a picture format of the N × M data. The method of claim 4, wherein Wherein the picture format is an interlaced scan field format or a sequential scan field format. The method of claim 4, wherein the scan pattern selection information is information for selecting a scan pattern that scans a horizontal high frequency component value relatively earlier than a vertical high frequency component value when the picture format is an interlaced field format. Decoding method characterized in that. The method of claim 4, wherein the scan pattern selection information is configured to scan vertical frequency component values and horizontal frequency component values in an equal order when the picture format is an interlaced field format or a sequential scan field format. Decoding method characterized in that the information to select a scan pattern. The method of claim 4, wherein step (c) (c1) inverse quantization of the N × M data; And (c2) transform-decoding the dequantized N × M data. In the method for decoding video data, variable length decoding the image data; (b) inversely scanning the variable length decoded image data using at least one scan pattern selected from a plurality of scan patterns to generate N × M data; And (c) source decoding the N × M data, And the at least one scan pattern is selected from among a plurality of scan patterns based on a picture format of the N × M data. The method of claim 9, And the picture format is an interlaced scan field format or a sequential scan field format. In the method for decoding video data, (a) variable length decoding the image data to obtain symbol data; (b) obtaining transform coefficient values from the symbol data; (c) generating scan pattern selection information for selecting a predetermined scan pattern among the plurality of scan patterns; (d) reverse scanning the transform coefficient values into N × M data using a scan pattern according to the generated scan pattern selection information; And (e) source decoding the inversely scanned data; And the scan pattern selection information is generated based on a picture format of the N × M data. The method of claim 11, Step (c) is (c1) analyzing a picture format of the N × M data; And and (c21) generating the scan pattern selection information for selecting a scan pattern that scans a horizontal high frequency component value relatively earlier than a vertical high frequency component value when the picture format is an interlaced field format. Way. The method of claim 11, Step (c) is (c1) analyzing a picture format of the received N × M data; And (c22) generating scan pattern selection information for selecting a scan pattern for scanning the vertical frequency component value and the horizontal frequency component value in equal order when the picture format is an interlaced field format or a sequential scan field format; Decoding method, characterized in that. In the apparatus for encoding video data, A first encoder which first encodes the image data to obtain N × M data; An encoding control unit providing scan pattern selection information for selecting a scan pattern among a plurality of scan patterns; And A second encoder which scans the N × M data using the scan pattern according to the scan pattern selection information, and encodes the scanned NXM data, And the scan pattern selection information is generated based on a picture format of the N × M data. The method of claim 14, And the picture format is an interlaced scan field format or a sequential scan field format. The method of claim 14, The encoding control unit And scan pattern selection information for selecting a scan pattern that scans a horizontal high frequency component value relatively earlier than a vertical high frequency component value when the N × M data is obtained from image data in an interlaced scan field format. The method of claim 14, The encoding control unit When the N × M data is obtained from image data in an interlaced field format or sequential scan field format, scan pattern selection information for selecting a scan pattern for scanning the vertical frequency component values and the horizontal frequency component values in equal order is provided. And an encoding device. An apparatus for decoding video data, A second decoder which decodes the image data to generate transform coefficient values; A scan pattern selection unit generating scan pattern selection information for selecting a predetermined scan pattern among a plurality of scan patterns; And A reverse scanning unit which reverse-scans the transform coefficient values into N × M data using the scan pattern according to the generated scan pattern selection information, And the scan pattern selection information is generated based on the size of the N × M data. The method of claim 18, And the picture format is an interlaced scan field format or a sequential scan field format. The method of claim 18, The scan pattern selector Scan pattern selection information for analyzing a picture format of the N × M data, and selecting a scan pattern for scanning a horizontal high frequency component value relatively earlier than a vertical high frequency component value when the picture format is an interlaced field format. And a decoding device. The method of claim 18, The scan pattern selector Scan pattern selection for analyzing the picture format of the N × M data and selecting a scan pattern for equally scanning the vertical frequency component value and the horizontal frequency component value in the case of the interlaced field format or the progressive scan field format as a result of the analysis. Decoding apparatus characterized by generating information.

Images