JP2001197503A - Coding method for digital interlace scanning moving picture - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a coding method for an interlace scanning moving picture that conducts error compensation with high performance and reproduces a moving picture with a high image quality. SOLUTION: The coding method for an interlace moving picture in the unit of pictures includes a 1st step where additional information in the unit of pictures is transmitted, a 2nd step where whether or not a coded video packet is a 1st packet is discriminated, a 3rd step where video packet header information including required additional information for interlace scanning decoding is transmitted when the video packet is the 1st packet, a 4th step where whether or not a picture coding type is an intra-picture is discriminated, a 5th step where an interlace scanning intra-picture video packet is transmitted when the picture coding type indicates the intra-picture, a 6th step where a video packet of an interlace scanning inter-picture video packet is transmitted when the when the picture coding type indicates no intra-picture, and a 7th step where when the video packet coded in the 5th or 6th step is not the final packet, the procedure returns to the 2nd step again and when the video packet coded in the 5th or 6th step is the final packet, the digital interlace scanning coding is completed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for encoding and decoding a digital interlaced moving image, and more particularly, to a method for compressing and transmitting an interlaced moving image when transmitting it. The present invention relates to a method of encoding and decoding interlaced scanning moving images that can reproduce high-quality moving images by minimizing degradation due to (transmission error or channel error).

[0002]

2. Description of the Related Art In general, digital moving pictures are classified into progressive video and interlaced video according to a method of constructing a signal. As shown in FIG. 6 (a), a forward scanning video signal is a frame structure or picture in which all signals in a video are acquired at the same time and scanned at the same time. [The term is mixed in this specification] is a signal.
As shown in (b), in the interlaced video signal, the signal in the video is divided into an upper field (top field) [solid arrow] and a lower field (bottom field) [dotted arrow]. This is a video signal having a field structure with different signal acquisition and scanning times. In the interlaced video signal, the upper field and the lower field are sequentially constituted by one line each, and the interlaced scanning frame is formed by interposing the lines of each constructed field (a solid arrow and a dotted arrow are intersected). Is configured.

In interlaced video, there is a time difference between the upper and lower fields, as shown in FIG. 6b. In FIG. 6b, the upper field is shown to precede in time, but in some cases, the lower field may precede in time. Due to the time difference between the two fields, the signal characteristics between adjacent lines in the jump frame may be different. In other words,
The signal characteristic of the ith line of the interlaced scanning frame is i + 1
It may be different from the signal characteristics of the i-th or (i-1) -th line. On the other hand, the signal characteristics and similarity (correlation) of the (i + 2) th line or (i−2) th line of the same field made at the same time
lation) can be relatively large. Such characteristics become more remarkable as the movement of an object existing inside the video screen becomes larger. Therefore, video coding devices developed with forward scanning video characteristics, such as motion estimation (motio)
n estimation) and motion compensation, the discrete cosine transform (DCT), etc., used for interlaced video coding, the coding efficiency (codin
g efficiency or coding gain) will decrease.

In order to eliminate such a decrease in coding efficiency, field-based motion estimation and motion compensation, adaptive field / fram adaptive (field / frame).
e) DCT and the like have been studied. In the present invention, the coding method developed based on the interlaced scanning video is generally called “interlaced coding method”.
d) ”, which is commonly referred to as“ progressive scan coding method ”.
ing method) ".

[0005] The interlaced scanning coding technique is an ISO / IE which is a global standardization organization related to a digital video coding method.
C JTC1 / SC29 / WG11 (International Standardization Or
ganization / International Electrotechnical Commissi
on Joint Technical Committee 1 / Sub Committee 29 / Wo
rking Group 11), MPEG-2 (Moving Picture Experts Group P) established for applications such as digital TV
hase 2) It is included in the moving picture coding standard and so on, and is therefore a technology often used in practically applied products.

On the other hand, digital video frames are defined and used in three types, such as I frame, P frame, and B frame, for high efficiency video compression and random access. Widely used. Here, information for classifying the I frame, the P frame, and the B frame is referred to as “picture coding type (picture coding type).
coding type) ". The characteristics of the above three types of frames are shown in FIG.
This will be described with reference to FIG. In FIG. 7, the upper arrow shows an example of the reference direction in the P frame, and the lower arrow shows the reference direction of the B frame. In this example, the interval between I-frames, the number of consecutive B-frames, etc., vary depending on the application or equipment.

First, I-frame (Intra coded frame)
Does not refer to other frames when encoding. Second, P
Frame (Predictive coded frame) used to be (previous)
Since the motion compensation prediction is performed with reference to the I frame or the previous P frame, and the coding is performed, coding with a high compression rate is possible. In this case, the I frame or the P frame to be referred to is referred to as a “reference frame”. The hue information (texture) is displayed on the time axis between successive frames of the video.
information) has a large degree of redundancy. Therefore, the motion information between the previous reference frame and the current P frame is estimated and the estimated motion information is compared with the method of encoding without reference like the I frame.
motion compensated prediction using (estimated motion information)
(motion compensated prediction), and the prediction error (pr
A method of encoding and transmitting ediction error and motion information is advantageous in terms of compression ratio.

Third, B-frame (Bidirectionally-pr
The edictive coded frame is the frame with the highest compression ratio.
Make predictions with reference to e). B frames use compensated prediction like P frames. However, the reference video (reference f
The highest compression ratio is used to select two better prediction performances among them. On the other hand, this frame does not serve as a reference image for another frame. Therefore, the previous frame is now the B frame, in display order.
(display order) closest previous I frame or P
Frame, and the subsequent frame is the next I frame or P frame closest to the current B frame in the display order.

When digital video is compression-encoded, I
Using frames, P frames, and B frames, the display order and coding order (coding orde
r) changes. Taking FIG. 7 as an example, the display order is I1, B1, P1, B2, P2, B
3, P3, B4, I2, but the encoding order is I1,
P1, B1, P2, B2, P3, B3, I2, and B4. Therefore, it is possible to decode and display the second frame B1 in the display order at the receiving stage after decoding the third frame P1. So, like a videophone,
Applications having a real-time video communication function require low delay coding, and such application products use only I frames and P frames without B frames to perform digital video. Is compressed and transmitted.

The example of FIG. 7 is provided for both forward scan video and interlaced video. However, in the case of the interlaced video, the point that two fields are interposed between the frames when the respective frames are formed is different from the case of the forward scan video as shown in FIG. 6B. Except for this point, the structure, prediction direction, and the like of the video are the same for both types of video. At the time of digital video compression encoding, motion estimation and compensation, hue information encoding, and the like are performed in units of a set of pixels having a certain size, and this is called a “macroblock (MB)”. The size of the most widely used macroblock is 16 pixels / line × 16 lines (hereinafter, 16 × 16).
).

[0011] The macroblock coding method is roughly classified into two methods, namely, "intra coding" and "inter coding". Intra-video encoding uses input hue information (texture
On the other hand, inter-video coding is a method of predicting hue information from a reference video and then coding the difference between the predicted information and the input information, that is, the error signal. I
Since a frame does not refer to another frame, all macroblocks in an I frame are subjected to intra-picture coding.
On the other hand, in the P frame, inter-picture coding is also possible, and generally, a coding method advantageous in terms of coding efficiency is selected from inter-picture coding and intra-picture coding. For example, if the motion estimation is incorrect, the prediction error becomes large, and encoding this may lower the encoding efficiency compared to the method of encoding the input video signal. In this case, the intra-picture coding method is more advantageous than the inter-picture coding method. Therefore, a macroblock to be intra-coded may exist in the P frame.

In summary, intra-picture coding and inter-picture coding are determined depending on whether the signal to be coded is an input hue signal or an error signal. In an I frame, all macroblocks perform only intra-picture coding, whereas in a P-frame, a macroblock to be intra-coded and a macroblock to be inter-coded can exist simultaneously.

In this specification, a macroblock to be intra-coded is referred to as an “intra macrobloc.
k) ", and the macroblock to be inter-coded is called an" inter macroblock ". Information indicating such a division is called a" macroblock type ". The video coding method includes a “frame-based coding” method for coding a rectangular frame, and an arbitrary form of area.
(object-based coding), which encodes only (arbitrary shape region).
The frame-based coding method is a moving image coding method in which the pixels of all frames are coded and transmitted, while the object-based coding method is a moving image coding method in which only the target area to be coded is coded and transmitted. Therefore, it is expected to play an important role in the future multimedia industry in general because it has an advantage that it can encode and transmit or manipulate only a user's desired object. Such an object-based coding method is based on the ISO / IEC JTC1
/ SC29 Adopted by MPEG-4, a standard established by WG11. In the object-based coding method, an encoder determines which pixel belongs to an object and which pixel belongs to a background among pixels in all frames.
(encoder) and a decoder (decoder) so that they can be distinguished identically.
encoding) and transmission.

FIG. 8 shows the M determined by the International Standards Organization.
1 shows a configuration diagram of a video encoder of PEG-4. FIG. MPEG
-4, "VOP (Vid
eo Object Plane) ". This coder uses information about the object (hue information and shape information) input to a motion estimation unit 13
The motion estimating unit 13 estimates the motion in macroblock units from the input information. In addition, the motion compensation unit (mot
ion compensation) 14 uses the motion information input from the motion estimation unit 13 to input hue information (textu).
re information) is motion compensated and predicted. Then, the signal predicted by the motion compensation unit 14 is input to a subtractor 16 together with the input signal, and a difference value is detected.
The difference value detected by the subtracter 16 is used as a hue information encoding unit (textu).
re coding) 18 to encode the hue information of the input object.

On the other hand, the motion compensation signal output from the motion compensation unit 14 and the hue information of the object encoded by the hue information encoding unit 18 are input to an adder 17 and added.
The output signal of the adder 17 is a previously reproduced VOP memory (previou
s reconstructed VOP memory) 15, which is stored and used for encoding the next input object. That is, the previous VO stored in the previously reproduced VOP memory 15
P is input to the motion estimator 13 and the motion compensator 14 and used for motion estimation and motion compensation.

The MPEG-4 moving picture encoder also performs intra-picture encoding and inter-picture encoding on a macroblock basis. Therefore, although not explicitly shown in FIG. 8, in the case of a macro block, since the motion compensation is not used, the input video signal is immediately input to the hue information coding unit, and the video signal reproduced here is also added to the adder. The data is input to the reproduction VOP memory immediately before without going through the like.

The shape information input together with the hue information is
The information is input to a shape information encoding unit (shape coding) 12, and the shape information is encoded. The outputs of the shape information encoding unit are a shape information bitstream and reproduced shape information. The reproduction shape information includes a motion estimation unit 13, a motion compensation unit 14, and a hue signal encoding unit 1 as indicated by a dotted line.
8 and can be used to encode motion estimation, motion compensation and object hue information on an object-by-object basis.
In the MPEG-4 video encoder, not only object-based coding but also frame-based coding is possible. In the frame unit coding, since shape information coding and transmission are unnecessary, as shown in FIG. 8, a shape information coding unit and its signal can be turned on / off by a switch. The signal that controls the switch is shape_coding_flag in FIG.

A motion information bitstream estimated and encoded by the motion estimator 13 (motion information bitstream)
A hue information bitstream (texture information bitstream) of the object encoded by the hue information encoding unit 18 and a shape information bitstream encoded by the shape information encoding unit 12
(shape information bitstream) is a multiplexer
r) After being applied to 19 and multiplexed, a buffer 2
0, and the multiplexed bit string (m
ultiplexed bitstream) is transmitted to the decoder side through a transmission medium. On the other hand, although not shown in FIG. 8, additional information in VOP units is also encoded and input to the multiplexing unit, multiplexed with other bit strings, and transmitted to the decoder.
The additional information in VOP units will be described later in more detail.

As described above, the V defined by MPEG-4
The OP corresponds to the frame of the existing standard draft, and the I VOP, P VOP, B VOP as well as the frame type.
Has a "VOP coding type". This VOP coding type differs from a rectangular frame only in that it has arbitrary shape information, and other characteristics (for example, motion compensation prediction direction) match.

As described above, during digital video compression encoding, motion estimation and compensation, hue information encoding, and the like are performed in units of macroblocks, which are a set of pixels of a fixed size.
When encoding an object having arbitrary shape information, a macroblock can be classified into three types in relation to the shape information.
First, there is a macroblock when included outside the object area. This macroblock has no object region inside, and thus does not require an encoding and transmission process. This macroblock is referred to herein as a "transparent macroblock". Second, when all the macroblocks are included in the object area, the technology developed in the existing frame-based coding can be directly used. In the present specification, such a macroblock is called an "opaque macroblock". Finally, there is a macroblock in which an object region and a non-object region coexist. In this specification, such a macroblock is referred to as a “boundary macroblock”.
k) ”. Such a macroblock is called a receiving end (recei
In order to perform decoding in ver), shape information indicating the object region and hue information in the object region should be encoded and transmitted. Of course, in a macroblock for which motion estimation and compensation is performed, such as an inter-picture macroblock in a P frame or a B frame, motion information should also be encoded and transmitted.

FIG. 9 shows a transparent macroblock (TMB), an opaque macroblock (OMB), and a boundary macroblock.
(BMB) shows an example. Here, hatched portions indicate object regions to be coded, and small squares indicate macroblocks. In this example, the number of macroblocks (TMB) included outside the object area is 6
Macroblock (OMB) included inside the object area
, And 22 macroblocks (BMB) in which the inside and outside of the object area exist simultaneously.

On the other hand, research for improving the coding efficiency of interlaced scan video having signal characteristics different from that of forward scan video has been actively conducted. Such interlaced scan coding methods are roughly classified into two types: field / frame adaptive DCT and field-based motion information estimation and prediction. As a typical example of the interlaced scan encoding method, MP
EG-4 Final Draft of International Standard (FDIS)
Only two of the methods described in are described briefly.

First, field / frame adaptive (ada
ptive field / frame) DCT. The final stage of the video compression encoding method is a DCT (discrete cosine transform). A 16 × 16 macroblock is divided into four blocks having a size of 8 × 8, and DCT is performed for each block. FIG. 10 shows the relationship between a macroblock and four blocks. In the case of a macroblock within a video, the original hue signal is subjected to DCT, and in the macroblock between video, a motion compensation prediction error signal is subjected to DCT. The higher the similarity between the signals of adjacent pixels, the higher the efficiency of such a DCT.

In the case of interlaced video, since the signals in the same field are generated at the same time, there is a higher probability that the similarity between the signals on the same field is higher than the signals on the directly adjacent lines. Such features are the largest features that are distinguished from forward-moving video, especially when moving objects are present in the image. Therefore, in the case of interlaced video, both the DCT between pixels in the same field and the DCT that does not consider the field as in the prior art are allowed, and both DTCs are used.
A method having a high compression efficiency is selected and encoded in CT. Then, a flag (f
lag) to the receiving end. This is called "adaptive field / frame DCT".
In this specification, performing DCT between pixels in the same field is referred to as “field DCT”, and a method not considering a field is referred to as “frame DCT”, and a flag indicating the DCT method is indicated by dct_type.

The field / frame adaptive DCT will be described in detail as follows. First, the similarity between adjacent lines is compared with the similarity between every other line. If the similarity between adjacent lines is large, frame DCT is performed, and if the similarity is small, field DCT is performed. The judgment formula for similarity comparison is as follows.

[0026]

(Equation 1)

In the above equation, i and j indicate the vertical and horizontal axis positions in the MB, respectively, and pi, j indicates the hue value of the pixel at the i and j positions. If the left side is larger than the right side as in the above relational expression, it means that the similarity between pixels in the same field is high. At this time, as shown in FIG. 11, a shuffling step ((a) → (b)) is performed in which the position of the line is deformed and the upper field and the lower field are reconfigured to be adjacent to each other. Since then
The macro block shown in FIG. 10B is divided into four blocks as shown in FIG. 10, and DCT (field DCT) is performed. what if,
If the right side is larger than or the same as the left side, it means that the similarity of directly adjacent pixels is high. Therefore, without performing the shuffling step, the macroblock of FIG.
0, divided into four blocks, and DCT (frame D
CT).

In the field / frame adaptive DCT, a flag dct_type indicating whether the macroblock is a frame DCT or a field DCT is 1 bit.
(bit) and transmitted to the receiving end. The decoder at the receiving end decodes dct_type from the received bit sequence,
DCT transform coefficients of four blocks (transform coefficien
t) is decoded to perform inverse DCT. Thereafter, it is determined from dct_type whether it is a field DCT or a frame DCT. Suppose field D
If it is CT, the obtained macroblock is subjected to inverse shuffling ((b) → (a)) as shown in FIG. 11 to obtain a final macroblock signal.

The second is a field unit of motion information estimation in a P frame. The most common existing motion information estimation method is to estimate motion information from a reference video in macroblock units, and perform motion compensation prediction from a reproduced signal of the reference video using this information. At this time, it is assumed that all pixels in one macroblock have the same motion information. Therefore, it is preferable to transmit one piece of motion information per macroblock to the receiving end (as will be described later, four pieces of motion information may be transmitted per macroblock).

In the case of forward scanning video, such an assumption can be applied. However, in the case of interlaced video, as described above, since the video generation time differs for each field, it is more accurate to assume that all the pixels in one macroblock have the same motion information. . Therefore, in the case of interlaced scanning video, estimating motion information on a field-by-field basis enables more accurate motion information estimation, and can reduce the motion compensation prediction error to obtain a coding gain. However, two pieces of motion information must be transmitted for each macroblock. Therefore, in general, a better method is selected by comparing the motion information estimation on a frame basis with the motion information estimation on a field basis. Then, the selection information flag (field_predi
ction) is represented by 1 bit and transmitted to the receiving end.

It should be noted that, as shown in FIG. 12, the motion compensation of the current upper field is possible from both the upper field and the lower field of the previous frame. Therefore, the information flag indicating the prediction direction (predictio
n_direction) should be transmitted together. The same applies to the lower field of the current frame. In summary, there are the following five types of information to be transmitted to the receiving end per macroblock for motion information estimation and prediction in field units. 1) field_prediction 2) high-order field prediction_direction 3) low-order field prediction_direction 4) high-order field motion information 5) low-order field motion information The decoder at the receiving end converts the received bit string into a field_pr
Decode ediction information. If the motion compensation prediction is performed on a field-by-field basis, information 2) to 5) may be decoded from the bit string, and the decoding may be performed according to the information.

Although the case of the P frame has been described above, the field prediction is performed in a similar manner also in the case of the B frame encoding. Since the present invention is not directly related to B frame encoding and decoding, a detailed description thereof will be omitted. On the other hand, a channel in which a transmission error may occur in a bit sequence generated by compression-coding a digital video signal may be generated.
When transmitting through a (mobile communication channel)-, an error may occur. When the bit string received at the receiving end is decoded by the transmission error, there is a risk that a reproduced image quality is significantly damaged. Research into minimizing problems due to transmission errors is being actively pursued, which mainly includes error detection, error localization, and data recovery.
covery). Such research is commonly referred to as "error resilient technology", and the bit string structure that reflects such technology is referred to as "error resilient bitstream".
That.

In the following, an error tolerance technique described in MPEG-4 FDIS that performs encoding on a VOP basis will be briefly examined. This technique is applied to both object-based coding and frame-based coding having arbitrary shape information. First, there is a resynchronization method. Resynchronization signal (resynchronization marker or resync marker)
Enables resynchronization between the bit stream and the decoder after an error has been detected. When an error occurs on a bit string, bits between the immediately preceding resynchronization point and the immediately following resynchronization point are generally ignored. If the resynchronization method could effectively localize or minimize the amount of data ignored by the decoder, it would improve the performance of other techniques related to error resilience (data recovery or error concealment). Can be done.

The resynchronization method of MPEG-4 is an ITU-T moving picture coding standardization scheme H.264. 261 or H.264. 26
3 is similar to the GOB (Group of Blocks) structure used. In this standard, a GOB is composed of one or several rows of macroblocks (rows). Then, when a new GOB is started in the encoding process, the GOB header (hea
der) on the bit string. This header is GOB
Contains a start code, which the decoder
Be able to know the position of OB. In addition to the start code, the GOB header stores information necessary for restarting the decoding process. The GOB scheme is a spatial resynchronization. In other words, when a specific macroblock position is reached during the encoding process, a resynchronization signal is inserted into the bit string. The problem inherent in this scheme is that when the encoding process is variable bit-rate encoding, the resynchronization signal cannot be located at a constant interval on the bit stream.

The technique adopted in MPEG-4 inserts a resynchronization signal into a bit string periodically to avoid such a problem. That is, a video packet (video packe
The configuration of t) —the set of bit strings between resynchronization signals—is based on the number of bits included in the video packet instead of the number of macroblocks. For example, after one macroblock is encoded, the accumulated number of bits of the video packet is given a threshold value (threshosho).
If (ld value) is exceeded, a new video packet starts with the next macroblock.

FIG. 13 shows an example of an error-resistant bit string structure using a resynchronization signal. Bit strings in MPEG-4 are resync marker, macroblock number,
The video packet is composed of a video packet header composed of a quant scale, a header extension code (HEC), and the like, and a video packet composed of actual macroblock information (macroblock data). The resynchronization signal marks the start of a new video packet. For this reason, this signal must be distinguished from all possible VLC codes and various start codes. The header information including the resynchronization signal is located at the start of the video packet. This header is indispensable for restarting the decoding process from the resynchronization signal closest to the position where the error occurs after the error occurs in the video packet, and the position of the first macroblock (macroblock number) of the video packet. ) And quantization parameters (quant scale).

[0037] The video packet header may include a "header extension code". The header extension code is used for additional information (overhead inform
ation). That is, when the information in the VOP is lost due to the transmission error, the decoding can be performed using the information in the VOP in the video packet. Here, typical additional information is VOP_coding_type, intra_dc_vlc_thr, VOP_f
Includes code_forward and VOP_fcode_backward. Where VO
P_coding_type indicates that the current VOP to be encoded is IVOP, P
Intra_dc_vlc_thr indicates information on whether or not to use a VLC table of AC (alternating current) information when encoding DC (direct current) information. . VOP_fc
ode_forward is the forward motion information estimation (forward motion est
VOP_fcode_backward is information indicating the size of the search area during backward motion estimation in B VOP. As described above, the information included in the header extension code is necessary information for the decoder to independently decode the video packet.

In MPEG-4, when using the error resilience encoding method, it is necessary to partially modify the compression encoding method. For example, all predictive coding methods (pre
dictive coding methods) are used for error propagation (error prop
In order to prevent aggregation, the information used for prediction should be limited to information in the same video packet. In other words,
During prediction (for example, AC / DC prediction, motion vector prediction (mo
option vector prediction)), VO
It should be considered the same as the P boundary.

Second, there is a data partitioning technique as an error resilience coding technique. The performance of an error concealment method, which is essential for a video decoder that is resistant to transmission errors, depends considerably on the resynchronization method. If the resynchronization method limits the errors fairly locally, the performance of the error compensation technique will be higher.

In MPEG-4, the necessity of an improved error localization method was recognized, and a new technique called data division was developed. This data division is, in the case of P VOP,
The information of all the macroblocks in the video packet is separated based on the hue information and the motion information, and the second resynchronization signal is inserted between the motion information and the hue information. At this time, the fact that the data division is used is notified to the receiving end as additional information so that the decoder can know. Motion division of the second resynchronization signal for data division to distinguish it from the existing resynchronization signal
Expressed as _marker.

FIG. 14 shows a bit string structure using data division. As can be seen from the figure, in the case of data division, the macroblock data in FIG.
motion and overhead information part and t
exture information part. motion_marke
If r is correctly decoded, it can be considered that the motion information of all the macroblocks in the video packet and the related additional information have been correctly decoded. If it is detected that a transmission error exists in the hue information bit string to be subsequently transmitted and decoded, all the hue information in the video packet is ignored, and only the motion information decoded before the motion_marker is used. Then, motion compensation prediction is performed from the reference VOP already decoded and stored in the memory to reproduce all macroblocks in the video packet.

In the P VOP, if the motion information is important information, in the case of the I VOP, the DC information is important information. Therefore, information of all macroblocks in a video packet is separated centering on DC information and AC information, and a resynchronization signal is inserted between them. In the case of the I VOP, since there is no motion information estimation and compensation, the motion information is not considered. If a transmission error occurs in the AC information, the AC information is ignored, and all the macroblocks in the video packet are reproduced using only the DC information. At this time, the resynchronization signal to be inserted is referred to as “dc_marke
r ”, the data division encoding method is
There is an advantage that information to be ignored can be reduced when a transmission error occurs. In the case of FIG. 13 which is not data division, if a transmission error is detected in one video packet, all information of the video packet is discarded. Then, a resynchronization signal after the occurrence of the transmission error is searched, and decoding is started again from that time. However, as described above, when data division is performed, in the case of a PVOP, when a transmission error occurs in a hue signal, there is an advantage that motion information can be used for reproducing a video signal without having to ignore it. I
In the case of VOP, when a transmission error occurs in an AC signal, there is an advantage that the DC information can be used for reproducing a video signal without having to ignore the DC information.

So far, the resynchronization signal and the data division technique have been described. The third error tolerant technology is reverse VLC
There is a way. Reverse VLC code (Reversible Variable Len
gth Codes (RVLC) method, the coding table is forward
It is designed so that decoding in both directions such as (forward direction) and reverse direction is possible. Since there is an error, decoding is performed in the backward direction from the resynchronization signal immediately after the error occurrence point for the bit string that cannot be decoded in the forward direction. Therefore, in this case, the number of bits ignored by the decoder due to transmission errors is reduced. Currently, in MPEG-4, RVLC is applied only to the encoding of DCT transform coefficients.

Of the error tolerance techniques described above, the data division coding scheme developed so far is difficult to apply to the interlaced scanning coding scheme. Since the development of the existing interlaced scanning coding method is assumed to be applied to an environment where transmission errors are small, such as Digital TV, an error resilience technique is unnecessary. However, when applications for encoding and transmitting interlaced scanning moving images are required in a mobile communication environment where transmission errors frequently occur, the interlaced scanning video encoding method causes deterioration in image quality due to transmission errors. This has the problem of becoming considerably large.

In the following, in order to more clearly describe such a problem, it is divided into an intra picture (I frame) and an inter picture (P frame or B frame).
The conventional video packet encoding method will be described in more detail. FIG. 15 is a flowchart of an existing intra-picture video packet encoding method. Assuming that the number of macroblocks present in this packet is NMB, first, all shape information of NMB macroblocks belonging to this packet, additional information for decoding DC information, and DC information are transmitted. . If the macroblock is a transparent macroblock having no object pixel, D
There is no need to transmit additional information and DC information for decoding C information. Then, it transmits dc_maker, and also transmits additional information and AC information for AC information of NMB macroblocks. dc_maker is a signal for separating DC information and AC information, and is a signal to be separated from all codes available in the entire encoder.

When such an existing intra-picture video packet encoding method is applied to interlaced scanning intra-picture encoding, many problems occur. For example,
When encoding interlaced video, field DCT is performed, and dct_type, which is information on the field DCT, is transmitted. However, existing methods do not consider such information transmission at all. Therefore, when the interlaced scanning intra-picture is transmitted by the video packet coding method of the existing method, it is impossible to appropriately reproduce the video by the decoder due to lack of information. That is, after reproducing DC, FIG.
The decoder cannot know which of the methods (a) and (b) is to be reproduced.

FIG. 16 is a flowchart showing a method for encoding a video packet of an existing inter picture. Assuming that the number of macroblocks present in this packet is NMB, first, all the shape information of NMB macroblocks belonging to this packet, additional information for motion information,
Then, the motion information is transmitted. Then, a motion maker is transmitted, and additional information and hue information for hue information of NMB macroblocks are also transmitted. Motion maker is a signal for separating motion information and hue information, and all codes that can be used by the overall encoder are signals to be separated.

As the additional information for motion information, a flag indicating that the coded bit is not generated at all for the macroblock, and in what mode the motion information is (for forward scanning, one motion information per macroblock is used). (There is a transmission mode and a mode for transmitting four pieces of information). The additional information for the hue information includes information indicating in which block the hue coefficient to be transmitted exists, a flag indicating whether or not to predictively encode the hue information AC coefficient.

When such an existing inter-picture video packet encoding method is applied to interlaced inter-picture encoding, there are many problems. For example, it is not necessary to partition each field when encoding a forward scan inter-picture, but it is necessary to partition each field when encoding interlaced video. Therefore, if the existing inter-picture video packet encoding method is applied to the interlaced scanning moving image as it is, there is a problem that a transmission error cannot be properly dealt with.

[0050]

SUMMARY OF THE INVENTION An object of the present invention is to provide a decoder capable of performing high-performance error compensation even when transmitting a compressed interlaced moving picture through an error-prone channel such as a mobile communication network. And a method of encoding an interlaced scanning moving image capable of reproducing a high-quality moving image.

[0051]

In order to achieve the above object, the present invention provides a first step of transmitting additional information in a picture unit, in encoding a skipped moving picture in a picture unit. A second step of determining whether the video packet to be used is the first packet of the picture;
If the result of the second step is not the first video packet,
A third step of transmitting video packet header information including additional information necessary for interlaced scanning decoding, a fourth step of determining whether the encoding type of the picture is an intra picture, and a fourth step. The fifth step of transmitting the interlaced scan intra-picture video packet when the encoding type of the picture is an intra picture, and the interlaced scan inter picture when the encoding type of the picture is not an intra picture in the fourth step. A sixth step of transmitting the video packet, and determining whether the video packet encoded in the fifth step or the sixth step is the last packet of the picture, and if not, proceeds to the second step again; On the other hand, if the packet is the last packet, the seventh step of terminating the digital interlaced scan encoding includes the following steps: The law provides.

[0052]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments according to the technical concept of the present invention will be described in detail with reference to the accompanying drawings. FIG. 1 is a flowchart illustrating a data division encoding process for encoding each picture of an interlaced moving image in video packet units according to the present invention.

The encoding method of the digital interlaced scanning image according to the present invention will be described in more detail as follows. When one picture of the interlaced scanning moving image to be coded is input, picture coding is started, and the first step is to transmit picture unit additional information necessary for picture coding (S1). Such picture unit side information may have some differences depending on the applied standard or application. In the present invention, such additional information is not particularly limited. For example, in the case of MPEG-4, examples of the picture unit side information are picture coding type, vop_coded, vop
_rounding_type, quantization interval size information, and the like. However, such additional information in picture units includes a picture coding type indicating whether the picture to be coded is an inter picture or an intra picture, and parameters of interlaced scan coding used for interlaced scan decoding. , Top_field_first, alternate_vertical_sca
n_flag etc. must be included. Where t
op_field_first is information indicating whether a higher-order field or a lower-order field precedes in time, and alternate_vertical_scan_flag is information indicating a method of scanning a transform coefficient after DCT.

The third step (S3) is a step of transmitting video packet header information including additional information necessary for interlaced scan decoding, and is located before each video packet. However,
Since the first video packet of each picture appears immediately after the additional information transmitted in the first step (S1), the first video packet does not transmit a video packet header. Such control is performed in the second step (S2). That is, the second stage
In (S2), it is determined whether the video packet to be encoded is the first packet of the picture, and a third step (S3)
Is controlled. If the video packet to be encoded is not the first packet of the picture in the second step (S2), the video packet header is transmitted in the third step (S3). The video packet header is information used when the picture-based additional information transmitted in the first step (S1) is lost due to a transmission error or the like, and thus includes the picture-based interlaced scanning additional information. First stage (S1)
Similarly to the above, the type of the detailed additional information is not limited in the present invention, but the interlaced additional information in the picture unit basically indicates whether the picture to be coded is an inter picture or an intra picture. picture codi
ng type and interlaced scan encoding parameters used for interlaced scan decoding (top_field_first, alternate_ver
tical_scan_flag).

In the fourth step (S4), it is determined whether the picture to be coded is an intra picture or an inter picture.
The step (S5) of transmitting the inter-picture video packet is performed. If the picture is an inter-picture, the step of transmitting the inter-scan video picture packet of the sixth step (S6) is performed. In the present invention, the reason why video packets are transmitted separately for the case of an intra picture and the case of an inter picture is that important information in both pictures is different.

Subsequent to the fifth step (S5) or the sixth step (S6), finally, it is determined whether or not the video packet after the coded transmission is located at the end of the picture. Proceeds to the second stage (S2) again to start a new packet, and if it is the last packet, ends the interlaced scan coding (S7). Hereinafter, the fifth step (S5) of transmitting the video packet of the inter-scanning intra picture in the method of encoding the inter-scanning moving image of the present invention will be described in more detail with reference to FIG.

At the time of video packet encoding transmission of an interlaced scanning intra picture according to the present invention, first, a shape information bit string obtained by encoding shape information of the macroblock is transmitted to the decoder side (S51). Then, a transparent macroblock (t
It is determined whether the block is a ransparent macroblock. If the block is a transparent macroblock, there is no need to send any signal, and the process proceeds to step 56 (S52). The encoding method of the present invention can be applied not only to object-based encoding but also to frame-based encoding. In frame-based encoding, transmission of shape information is unnecessary, and therefore, the fifty-first step of shape information transmission is performed.
(S51) and the fifth step of determining and processing whether it is a transparent area
The two steps (S52) are omitted.

In the 52nd step (S52), if it is not a transparent macroblock, various additional information necessary for decoding DC information of the macroblock is transmitted (S5).
3). Such additional information includes macroblock type,
It may include information such as the quantization interval size. In the present invention, information to be transmitted at this stage is not limited to specific information, except that additional information necessary for decoding DC information must be transmitted.

In the present invention, after transmitting the additional information for decoding the DC information, a flag (dct_type) indicating whether the DCT is performed in the frame mode or the field mode for each block of the macroblock is transmitted. (S5
4) Do it. This is to check the reproduced dct_type in the decoder, perform inverse quantization and Inverse DCT, and know whether to perform shuffling on the reproduced result. Following the fifty-fourth step (S54), the fifth
In the fifth step (S55), the DC information of the block of the macro block is transmitted. In a 56th step (S56), it is determined whether the transmitted macroblock is the last macroblock of the video packet. If the transmitted macroblock is the last macroblock, the 57th step (S57) is performed. Step 51 for macro block information (S51)
Resume.

In the present invention, there are various determination methods in the 56th step (S56). For example, the length of a video packet can be specified by a fixed number of macroblocks (such as H.263 in a GOB structure), or when the accumulated number of generated bits exceeds a predetermined threshold value, Is completed (like MPEG-4). In the present invention, the method for determining the last macroblock in one video packet is not limited to a specific method.

In step 56 (S56), if the packet is the last packet, a marker signal dc_marke for separating DC information and AC information of all macroblocks in the video packet.
r is transmitted (S57). It must be distinguishable from all codes available for all video coding systems in which the coding method of FIG. 2 is used. For example, MPE
In G-4, a bit string of “110 1011 0000 0000 0001” is used as dc_marker.

Next, in step 58 (S58), the decoder for all the macroblocks included in the video packet transmits the additional information used in the process of decoding the AC information. As additional information used in the process of decoding AC information, coded blockpattern (signal indicating whether there is an AC coefficient to be transmitted in each block in a macroblock) and ac_pred_flag (predictive coding of AC information)
coding) signal). Finally, in step 59, the AC information of all macroblocks included in the video packet is transmitted.

FIG. 2 shows that all macroblock information in the video packet is transmitted in each of the 58th step (S58) and the 59th step (S59).
However, step 58 (S58) and step 59 (S59) may be performed on each macroblock in the video packet, and this may be performed on all macroblocks in the video packet.

In the method of transmitting an intra-picture video packet according to the present invention, step 54 (S54) shows the most important features of the present invention. If the dc_marker transmitted in step 57 (S57) is correctly decoded and a channel error is detected from the transmitted AC information,
The information is ignored, and the video packet is reproduced using only the DC information transmitted before dc_marker. At this time, dct_
If the type is not transmitted, the decoder does not know whether the macroblock is in the field mode or the frame mode, and the image quality is greatly deteriorated. However, when the dct_type is transmitted as in the present invention, the decoder at the receiving end determines whether the dct_type is a field DCT or a frame DCT.
T is determined, and if it is a field DCT, reverse shuffling is performed to reconstruct the last macroblock signal (see FIG. 11). As compared with the method of the first aspect, it is possible to minimize the influence of the generated channel error and to decode a high-quality reproduced video.

FIG. 3 shows the step (S6) of the inter-scan inter-picture video packet transmission in the inter-scan video coding method according to the present invention. The method of transmitting a video packet of an interlaced inter-picture according to the present invention will be described in detail. First, in the case of an object-based decoder, a shape information bit string obtained by encoding shape information of a macroblock is transmitted to a decoder side. Next, it is determined whether the macroblock is a transparent macroblock containing no object pixel (S61). If the macroblock is a transparent macroblock, no signal needs to be sent. ) (S62). Although the method of the present invention can be applied to frame-based coding, the frame-based coding does not require the transmission of shape information.
1) and a 62nd process of determining and processing whether it is a transparent area
Step (S62) is omitted.

In the 62nd step (S62), if it is not a transparent macroblock, various additional information necessary for decoding the motion information of the macroblock is transmitted (S63). Such additional information is information indicating coding (information indicating whether there is information to be transmitted in the macroblock.
coded), macroblock type, quantization interval size, and the like. In the present invention, such additional information is not limited to a set of specific information, but among the information transmitted in step 63 (S63), whether the macroblock is an inter macroblock using motion compensation prediction, or Macroblock type information indicating whether an intra macroblock does not use compensated prediction is always included. After the transmission of the additional information for the motion information, the macroblock type is checked. If the macroblock is an intra macroblock, the motion information transmission is unnecessary.
The step 69 (S69) is performed (S64). Stage 64
In step (S64), if the macroblock type is an inter macroblock, information field_prediction indicating whether the macroblock performs frame-based motion compensation prediction or field-based motion compensation prediction is transmitted ( S65). In step 66 (S66), this mode is determined, and if motion compensation prediction is to be performed on a field basis, field-based movement information is transmitted (S6).
7) If the motion compensation prediction is performed on a frame basis, frame-based motion information is transmitted (S68). In the example shown in FIG. 3, when "field_prediction" has a value of "1", it indicates that the mode is the field mode. Stage 67
In (S67), the field-based motion information is, as described above, the motion information of each field and the prediction direction of each field.
Including prediction_direction. Step 68 (S68)
As the frame-based motion information, basically, a mode of transmitting one motion information per macroblock or a mode of transmitting four motion information (one motion information per block) per macroblock can be used. After transmitting the field-based or frame-based movement information, it is determined whether the transmitted macroblock is the last macroblock of the video packet. If the transmitted macroblock is the last macroblock, step 70 (S70) is performed. If not, the 61st step (S
61) is restarted. In the 69th step (S69), there are various methods for determining whether the block is the last macroblock. For example, the length of a video packet can be specified by a fixed number of macroblocks (such as H.263 in a GOB structure), or when the accumulated number of generated bits exceeds a predetermined threshold value, It can be determined that the video packet ends (like MPEG-4 video packet information). In the present invention, the method for determining the last macroblock in one video packet is not limited to a specific method.

If it is the last packet in step 69 (S69), a marker signal motion_marke for separating motion information and hue information of all macroblocks in the video packet.
r is transmitted (S70). This must be able to be distinguished from all codes usable in the entire moving picture coding system using the coding method of FIG. 3-all variable length codes, signals for various resynchronizations, and start signals. . For example, in MPEG-4, "1 1111 0000 0000 0001"
Is used as motion_marker.

Next, in a 71st step (S71), in all macroblocks included in the video packet, additional information used in a process of decoding hue information by a decoder is transmitted. For example, coded bl explained in FIG.
ock pattern, ac_pred_flag, quantization interval, and dct_
type. Finally, in the 72nd step (S72), DC information and AC information, that is, hue information, are transmitted in all macroblocks included in the video packet. As in FIG. 2, for each of the 71st step (S71) and the 72nd step (S72) of FIG. 3, it is indicated that information of all macroblocks in the video packet is transmitted. However, step 71 (S71) and step 72 (S71) are performed for each macroblock in the video packet.
72), and this may be performed for all macroblocks in the video packet.

Of the steps shown in FIG. 3, the 65th step
Step (S65) and step 67 (S67) show the most important features of the present invention. Assuming that the moti transmitted in step 70 (S70)
If on_marker is correctly decoded and a channel error is detected from the subsequently transmitted hue information, all hue information of the video packet is ignored,
A video packet is reproduced using only motion information transmitted before motion_marker. At this time, field_predicti
on (S65) and transmission of motion information in field units (S6
7) allows the decoder to know whether the macroblock is in the field mode or the frame mode. When the macroblock is in the field mode, the motion information can be accurately decoded in field units. Therefore,
Even if hue information is lost due to a channel error, it is possible to accurately perform motion compensation prediction on a coded field basis, thereby minimizing the influence of the channel error,
High-quality reproduced video can be decoded.

FIGS. 2 and 3 show flowcharts of the video packet coding method of the inter-picture and the inter-picture of the interlaced scanning moving picture proposed by the present invention, respectively. On the other hand, these figures show the information transmission order as in FIG. Therefore, each decoder shows the same flowchart. However, substituting “reception” for “transmission” in FIGS. 2 and 3 results in a flowchart of a video packet decoding method for an intra-picture and an inter-picture of an interlaced moving image. Therefore, the detailed description thereof is omitted in the present invention, and the flow charts of the video packet decoding method for the inter-scan moving picture intra picture and the inter picture according to the present invention are shown in FIGS. 4 and 5, respectively.

The decoding method of FIGS. 4 and 5 is basically the same as that of FIG.
Or, it has the same flow as in FIG. 3, but the control method of the decoder is slightly different. As described above, in the encoding methods of FIGS. 2 and 3, since the encoder knows the length of the video packet, if the specific condition is satisfied, dc_marke
Transmit r or motion_marker. However, since the condition cannot be known in the decoder, a process of examining the next bit string is necessary. More specifically, in the case of decoding the intra-picture video packet shown in FIG. 4, after receiving DC information, first confirming whether or not the next bit string is dc_marker, if not, receiving the shape information of the next macroblock if not dc_marker If the bit string is dc_marker, the process of receiving dc_marker is performed. Similarly, in the case of video packet decoding of the inter-picture of FIG.
After receiving the motion information, the next bit string is motion_marke
After first checking whether it is r, if it is not motion_marker,
Receiving the shape information of the next macroblock,
When the motion_marker is confirmed, motion_mark
Perform the process of receiving er.

[0072]

As described above, according to the present invention, when encoding an interlaced scanning moving image, it is encoded using a data division encoding method, and the information of the encoded video packet is encoded according to the present invention. In the case where the interlaced scanning moving image is transmitted through a channel where a large number of transmission errors occur, the signal is transmitted to the decoder side according to the transmission order proposed in the above, and the decoder side performs the decoding in the order proposed in the present invention. This also provides an effect of minimizing image quality degradation due to channel errors on the decoder side.

[Brief description of the drawings]

FIG. 1 is a flowchart of a picture unit encoding method of an interlaced scanning moving image according to the present invention.

FIG. 2 is a flowchart of a method of encoding a video packet of an interlaced intra picture according to the present invention;

FIG. 3 is a flowchart of a video packet encoding method of interlaced inter-picture according to the present invention;

FIG. 4 is a flowchart of a method for encoding a video packet of an interlaced intra-picture according to the present invention;

FIG. 5 is a flowchart of a video packet decoding method of interlaced inter-picture according to the present invention;

FIG. 6 is a transmission flow chart on a time axis of a general forward scanning video and an interlaced video.

FIG. 7 is a diagram illustrating an example of picture (frame) coding types (I frame, P frame, and B frame) and the prediction direction of each type of frame.

FIG. 8 is a configuration diagram of an MPEG-4 moving picture encoder which has been determined by the International Standards Organization.

FIG. 9: TMB (transparent macroblock), OMB (opa
FIG. 4 is an exemplary diagram illustrating a que macroblock) and a BMB (boundary macroblock).

FIG. 10 is an exemplary diagram showing one microblock divided into four blocks.

FIG. 11 is an exemplary diagram of a frame MB (a) and a field MB (b).

FIG. 12 is an exemplary diagram illustrating a direction of referring to a previous field when performing field prediction.

FIG. 13 shows an error resilient bitstream using a resynchronization signal.
FIG.

FIG. 14 is a structural diagram of an error-resistant bit string to which a data division method is applied.

FIG. 15 is a flowchart of a conventional video packet encoding method for forward scanning intra pictures.

FIG. 16 is a flowchart of a conventional video packet encoding method for a forward scanning inter picture.

[Explanation of symbols]

 Reference Signs List 12 encoding unit 13 motion estimating unit 14 motion compensating unit 15 previously reproduced VOP memory 16 subtractor 17 adder 18 hue information encoding unit 19 multiplexing unit 20 buffer

Continuing on the front page (72) Inventor Sentence Pillars 610 Modern apartment 603-2206 F term (reference) 5C059 KK01 LA05 MA05 MA23 MC02 MC32 MC34 ME01 NN01 PP04 PP14 RB02 RB09 RB14 RC12 RC26 RF09 SS06 TA46 TA73 TB04 TC18 TC27 TD13 UA02 UA05

Claims (13)

[Claims] In encoding a skipped moving picture in picture units, a first step of transmitting additional information in picture units and determining whether a video packet to be encoded is the first packet of the picture. The second stage of judging,
If the result of the second step is not the first video packet,
A third step of transmitting video packet header information including additional information necessary for interlaced decoding, and whether the video packet to be encoded in the second step is the first packet of the picture or the third step After transmitting the video packet header information, the fourth step of determining whether or not the coding type of the picture is an intra picture; and, if the coding type of the picture is an intra picture in the fourth step, A fifth step of transmitting an intra-picture video packet, and a sixth step of transmitting an inter-scan inter-picture video packet if the coding type of the picture is not an intra picture in the fourth step;
It is determined whether the video packet encoded in the fifth step or the sixth step is the last packet of the picture. If the video packet is not the last packet, the process proceeds to the second step again. And a seventh step of terminating the scan coding. 2. The additional information in a picture unit in the first step is information indicating whether a picture to be coded is an inter picture or an intra picture, and which of the upper and lower fields is temporally different. 2. The method according to claim 1, further comprising information indicating whether or not the scanning is performed prior to the scanning, and information regarding a method of scanning a transform coefficient after the discrete cosine transform (DCT). 3. The fifth step is a step of transmitting additional information for decoding DC information of the macroblock.
(S53), a fifty-fourth step (S54) of transmitting a flag indicating whether the macroblock is in the frame mode or the field mode (S54),
Step 55 of transmitting information (S55); determining whether the macroblock is the last macroblock of the video packet;
The 56th step (S56) proceeding to (S53) and the 56th step
If it is the last packet in (S56), a 57th step (S57) of transmitting a marker signal for separating DC information and AC information of the macroblock of the video packet and decoding of AC information of all macroblocks of the video packet The method according to claim 1, further comprising a fifty-eighth step (S58) of transmitting additional information for conversion, and a fifty-ninth step (S59) of transmitting AC information of all macroblocks of the video packet. The encoding method of the digital interlaced scanning moving image of the description. 4. The method of claim 5, wherein the transmitting of the additional information for decoding the DC information of the macroblock for the object-based video encoding is performed before the 53rd step (S53). A 51st step (S51) of transmitting
It is determined whether the macroblock to be encoded is a transparent macroblock. If the macroblock is a transparent macroblock, the process proceeds to step 56 (S56).
The method according to claim 3, further comprising a 52nd step (S52) proceeding to 3 steps (S53). 5. The digital signal as claimed in claim 3, wherein in the 53rd step (S53), the additional information for decoding the DC information includes information on a macroblock type and a quantization interval size. An encoding method for interlaced scanning moving images. 6. In the 56th step (S56), when the length of the video packet is specified by a fixed number of macroblocks or when the accumulated number of generated bits exceeds a predetermined threshold value, the video packet is 4. The method as claimed in claim 3, further comprising the step of determining the end. 7. In the 58th step (S58), the additional information for decoding the AC information is obtained by predictively encoding a signal indicating whether there is an AC coefficient to be transmitted in each block in the macroblock and the AC information. 4. The method according to claim 3, further comprising a signal indicating whether predictive coding has been performed. 8. The step (S63) of transmitting the additional information for decoding the motion information in units of macroblocks necessary for compounding the macroblock (S63).
And a 64th step (S64) of determining whether the macroblock is an intra macroblock or an inter macroblock, and if the macroblock is not an intra macroblock in the 64th step (S64), the motion compensation prediction of the macroblock Step 65 of transmitting information indicating whether the mode is the frame mode or the field mode (S6)
5), step 66 (S66) of determining whether the motion compensation prediction mode of the macroblock is the frame mode or the field mode, and if the field mode is selected in the step 66 (S66), the field mode Step 67 (S67) of transmitting the motion information of
In the case of the frame mode in step 66), the 68th step (S68) of transmitting the frame unit movement information and the 64th step (S68)
64) whether it is an intra macro block or
After transmitting the movement information in step S67 or step 68 (S68), it is determined whether the macroblock is the last macroblock of the video packet. If not, the process proceeds to step 63 (S63). Step 69 (S6
9) and in the 69th step (S69), if the current macroblock is the last macroblock, the seventh signal for transmitting a marker signal for separating motion information and hue information of the macroblock of the video packet is transmitted.
0 (S70), a 71st step (S71) of transmitting additional information necessary for a hue information decoding process in all macroblocks of the video packet, and a hue information of all macroblocks of the video packet. 3. The method according to claim 1, further comprising a 72nd step of transmitting (S72). 9. The sixth step (S61) of encoding the shape information of the macroblock and transmitting the encoded information to the decoder before the 63rd step (S63); It is determined whether the block is a block, and if the block is a transparent macroblock, the process proceeds to step 63 (S63),
On the other hand, if it is not a transparent macroblock, step 69 (S
The method according to claim 8, further comprising a 62nd step (S62) proceeding to 69). 10. In the 63rd step (S63), the additional information for coding motion information is information (not_coded) indicating whether there is information to be transmitted in the macroblock, and the macroblock is an inter macroblock. 9. The method according to claim 8, further comprising macroblock type information indicating whether there is a macroblock or an intra macroblock, and quantization interval size information. 11. The digital interlaced scanning moving image according to claim 8, wherein in the 67th step (S67), the field unit motion information includes motion information of each field and prediction direction information of each field. Encoding method. 12. The 69th step (S69) is to end the video packet if the length of the video packet is specified by a fixed number of macroblocks or if the accumulated number of generated bits exceeds a predetermined threshold value. 9. The method according to claim 8, further comprising the step of: 13. In the 71st step (S71), additional information for decoding hue information includes a signal indicating whether there is an AC coefficient to be transmitted in each block in the macroblock,
A signal indicating whether information has been subjected to predictive coding, a quantization interval, and a flag indicating a DCT method (dct_typ
9. The method according to claim 8, further comprising the step of: e).