JP2755851B2 - Moving picture coding apparatus and moving picture coding method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving picture coding apparatus and a moving picture coding method for coding a moving picture signal such as a video signal in video communication, transmission and broadcasting.
In particular, a moving picture coding apparatus using motion compensated prediction coding and
The present invention relates to a moving picture coding method .

[0002]

2. Description of the Related Art TV telephones, TV conferences, optical disk devices,
In a device such as a VTR and a CATV, a technique for encoding a moving image signal is required. As such a moving image encoding method, a pixel value of an image to be encoded (hereinafter, referred to as an encoded elephant image) is calculated by using a pixel value of an already encoded reference image designated by a motion vector. So-called motion-compensated predictive coding for predicting and coding the prediction error and the motion vector is known.

When this motion compensated prediction coding is applied to an interlaced image (field image) such as a normal television video signal, a motion accuracy equal to or more than 1/2 line accuracy within a field (a motion equal to or more than 1 line accuracy within a frame) is used. (Accuracy) cannot be handled because there is no corresponding pixel value in the reference image.

Therefore, a method has been proposed in which pixel values of pixels having no corresponding pixel on a reference image are interpolated using adjacent two-field images to perform motion compensation (for example, the 1990 Image Symposium on Image Coding). (PCSJ9
0), 8-1, "Study of adaptive line interpolation inter-field motion compensation method"). In this motion compensation method, an encoding target image signal is encoded using a reference image signal and an optimal motion vector. The reference image signal is created by interpolation from a signal at a position specified by a motion vector obtained by a motion vector search circuit among encoded image signals of the past two fields. Specifically, three field memories are prepared, and a signal created by intra-field interpolation from the output of the first field memory and the output of the second field memory are mixed at a mixing ratio km: 1-km. Generates a reference image signal. km varies with the amount of motion detected by the motion detection circuit based on the outputs of the first field memory and the third field memory.

FIG. 2 is a diagram showing the relationship between image signals according to the prior art. The relationship between the image signals 41, 42, 43 and the encoding target image signal 44 stored in the three field memories is as shown in the figure, and the vertical component of the motion vector is n + 1/2 lines (n
Is an integer), the corresponding pixel value in the image signal 41 is output as it is as the reference image signal. For example, when n = 0, the pixel value of the reference image signal with respect to the pixel value 45 of the encoding target image signal 44 is 46. If the vertical component of the motion vector is n lines in the field, the image signal 4
Interpolation value て using signals て adjacent to 2,43 interpolation values △
Create That is, for example, an interpolation value 49 serving as a reference image signal for a pixel value 45 in the encoding target image signal 44 is
The sum of the value obtained by multiplying the average value of the pixel values 46 and 47 by km and the value obtained by multiplying the pixel value 48 by (1-km), which is created by intra-field interpolation, is obtained. At this time,
If the motion is determined to be large by the motion detection circuit,
It is appropriate to interpolate the interpolation value △ mainly by using the signal adjacent to the interpolation value 画像 in the image signal of the same field as the interpolation value か ら. Therefore, if km is large and the motion is determined to be small, the interpolation is performed. Since it is appropriate to interpolate the value △ mainly by using the signal adjacent to the interpolation value 画像 in the image signal of the field adjacent thereto, km is set to be small.

As described above, by forming an interpolation value by adapting to the amount of motion using the image signals of two adjacent fields, a half of the field image can be formed.
An appropriate reference image signal can be generated for a motion accuracy equal to or higher than the line accuracy (a motion accuracy equal to or more than one line accuracy in a frame), and highly accurate motion compensation prediction can be performed.

However, in this method, it is necessary to detect the motion in the reference image as described above, and a motion detection circuit for that purpose is required. Also, in order to enable motion detection, three adjacent fields must be already coded. Therefore, when three adjacent fields have not been encoded, a problem arises in that motion detection cannot be performed.

FIG. 3 shows an example in which three adjacent fields have not been encoded. In this example,
Special playback on VTR (random access, fast forward playback,
In consideration of adaptation to a storage system that requires reverse playback or the like, two adjacent fields (fields 1, 2, 7, and 7,
8) is encoded in advance, and the remaining four fields (3 to 6 in the example in the drawing) are motion-compensated and predictive-encoded using two neighboring fields as reference images in advance. In such a case, motion detection cannot be performed because three adjacent fields have not been encoded.

On the other hand, in a moving picture coding apparatus using the above-mentioned motion compensation predictive coding method, conventionally, when searching for a motion vector for forward or backward motion compensation, a reference picture for searching for a motion vector is used. When the encoding target image is a non-interlaced image, it is limited to only one encoded image.
There has been a problem that accurate motion compensation cannot be performed on a moving image in which a movement in units of / 2 pixels exists. When an image is an interlaced image, two consecutive encoded images are used as described above. However, conventionally, since the same area is searched for both images, the amount of calculation for the search is extremely small. Had increased.

[0010] Among the above-mentioned moving picture coding methods, particularly, as a coding method for a storage medium such as a VTR or an optical disk, a moving picture coding method targeting a transmission rate of about 1 to 2 Mbps. Is MPEG
It is being standardized under one standard name. This is a method based on motion compensation inter-frame prediction + DCT.

At present, studies are underway on a scheme for encoding a moving image having a quality equal to or higher than that of a TV broadcast at a rate of about 2 to 10 Mbps for a similar purpose. In order to adapt the encoding method of MPEG1 to this, MPEG1 assumes a non-interlaced image as an input image, whereas a TV signal is a signal of an interlaced image. New ideas are needed. In encoding an interlaced image, inter-field / inter-frame adaptive prediction is known. This is because the field of interest and the scan phase coincide with each other (an odd field if the field of interest is an odd field, and an even field if the field of interest is an even field) and a field that does not match but is temporally close (for example, if the field of interest is In the case of an odd field, an even field adjacent thereto is switched, and when the field of interest is an even field, an odd field adjacent thereto is switched as a prediction signal. In recent studies, interpolation prediction that averages and predicts a motion-compensated signal for a plurality of fields has been studied (for example, F. Wang et al., "High-quality co-
ding of the even fields based on the odd fields of
interlaced videosequences ", IEEE trans.CS).

However, in performing predictive encoding using past fields and future fields as in MPEG1 encoding, even in an interlaced image,
It is desirable to incorporate a prediction method using both even and odd fields suitable for interlaced images. In this case, when a motion vector is sent to each field, the information amount of the motion vector increases, so it is necessary to reduce the information amount of the motion vector without reducing the efficiency as much as possible. That is,
It is required to improve the prediction accuracy for interlaced images to reduce the information amount of the encoded output of the prediction error, and to minimize the increase of the motion vector information. Technology has not yet been proposed.

[0013]

As described above, in the prior art, when a pixel at a point where no pixel exists on the reference image is interpolated by using an image of two fields adjacent to the reference image, the motion of the reference image is reduced. Since detection is required, a motion detection circuit is required and hardware becomes complicated, and there is a problem that motion cannot be detected if three adjacent fields have not been encoded.

In the prior art, when the image to be encoded is a non-interlaced image, the reference image is limited to only one encoded image when searching for a motion vector.
There is a problem in that accurate motion compensation cannot be performed for a moving image in which movement of a half pixel exists between adjacent images, and the coding efficiency is reduced. In this case, the amount of calculation for the motion vector search becomes very large, so that there is a problem that the motion vector search time becomes long and the hardware circuit scale increases.

Furthermore, the conventional techniques cannot effectively improve the prediction accuracy for an interlaced image, and have a problem that the amount of information of a motion vector sent to each field increases.

[0016]

[0017]

SUMMARY OF THE INVENTION It is an object of the present invention to provide a moving picture coding apparatus capable of improving the prediction accuracy of an interlaced picture to reduce the information amount of a coded output of a prediction error and minimizing the increase of motion vector information. And a moving picture coding method.

[0019]

In order to achieve the above object, the present invention provides first and second motion vectors between images of a continuous first and second reference field and an image of a field to be encoded. Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the following formula, and calculating error information of the prediction signal for the image signal of the image of the encoding target field. A mode in which encoding is performed, and the magnitude of the difference between the vector closest to the first motion vector and the second motion vector among a plurality of motion vectors that can be the second motion vector is smaller than or equal to a small value. The mode is used only when the mode is used. In other words, the present invention is a continuous first,
From the reference signals on the images of the first and second reference fields specified by the first and second motion vectors, respectively, between the image of the second reference field and the image of the current field, A mode for generating a prediction signal for the image of the target field and encoding error information of the prediction signal for the image signal of the image of the coding target field, wherein a plurality of motion vectors that can be the second motion vector Among them, there is a mode that is used only in a range where the magnitude of the difference between the vector whose direction is closest to the first motion vector and the second motion vector is a minute value or less. Further, in the mode, a difference vector corresponding to a difference between the vector closest to the first motion vector and the second motion vector among a plurality of motion vectors that can be the second motion vector is generated. The error information, the information on the first motion vector, and the information on the difference vector are each encoded. According to another aspect, the present invention is the first motion vector first reference field image on the reference signal to be more specified in Le between the image of the image and the encoding target field of the first reference field And a motion vector between the image of the second reference field and the image of the encoding target field that are continuous with the first reference field, wherein the image signal and the encoding target In the field
Said first of said second motion only when the magnitude of the difference in the direction closest vector to the motion vector exists second motion vector that Do the following minute value between the image vector
Obtains a reference signal on the image of the second reference field specified by Le, generates a predicted signal for the encoding target image field from the first and the image on the reference signal of the second reference field, the coding It is characterized in that error information of a prediction signal with respect to an image signal of an image of a target field is encoded. Furthermore, "the closest vector is the direction in the first motion vector among the plurality of motion vectors that can be a second motion vector" is, in other words, obtained by converting enlarged converted or reduced to "first motion vector Said
The image of the second reference field and the image of the encoding target field
Said second reference closest in direction to the vector to the image
Between the image on the field and the image of the encoding target field
Between the motion vector "and" the first motion vector between the image of the second reference field and the image of the encoding target field.
The direction closest to the vector obtained by replacing
The image of the second reference field and the image of the encoding target field
First motion vector in accordance with the ratio of the motion vector "or" coded field and a first time interval between the reference field and a coded field time interval between the second reference field between the image obtained in terms of size
The image of the second reference field to be encoded
Field whose direction is closest to the vector between
The image of the second reference field and the image of the encoding target field
Motion vector between the image ".

[0020]

[0021]

[0022]

[0023]

[0024]

[0025]

As described above, according to the present invention, the first and second motion vectors are generated between the images of the continuous first and second reference fields and the image of the encoding target field, and the first and second motion vectors are generated. First and second specified by the second motion vector, respectively.
By generating a prediction signal for the image of the encoding target field from the reference signal on the image of the reference field, efficient and accurate prediction suitable for an interlaced image can be performed. In a moving region in an image, a highly accurate prediction signal can be obtained, and the accompanying increase in motion vector information can be minimized by adopting a difference vector. It can be minimized.

Therefore, in the present invention, for each motion vector candidate at the time of motion vector search, it is assumed that those motion vector candidates correspond to the actual motion amount in the reference image, and the interpolation corresponding to the motion amount is performed. A value is generated from an interpolation value generation circuit, and a reference image is configured using the interpolation value.
Search for a motion vector. By doing so, the motion vector is finally searched based on the reference image using an appropriate interpolation value without performing the motion detection on the reference image. Becomes possible.

Further, in the present invention, when searching for a motion vector using an encoded two-screen image as a reference image, a search is performed over a wide range for an encoded image that is temporally close to the image to be encoded. By searching a coded image that is temporally far from the current image to be coded in a more limited narrow range, accurate motion compensation can be performed while suppressing the amount of calculation for motion vector search.

For example, by searching for an encoded image temporally close to the image to be encoded, at least one candidate of a motion vector with good encoding efficiency is obtained, and an operation such as extension of the vector is performed based on the motion vector. If it is performed, it is possible to limit the center of a region that is a candidate for a motion vector with high encoding efficiency between the encoding target image and the encoded image that is temporally separated. In this case, if a region selected as a motion vector candidate in a temporally encoded image is encoded using a motion vector between the temporally separated encoded image, the time immediately It is possible to limit the range in which a region that is a candidate for a motion vector between an image that is far apart and an image to be encoded exists.

The motion between images is very large, and there is no motion vector that can be a candidate even in an encoded image that is close in time, or the motion vector obtained as a candidate in an encoded image that is close in time is extremely low. If it is possible to predict that there is no motion vector with high coding efficiency in a coded image that is far apart in time, it is not necessary to search for a motion vector in a coded image that is far apart in time. , And the search range can be further limited. By doing so, a motion vector that requires at least a large amount of computation for searching and that has good coding efficiency is obtained.

On the other hand, when the motion between images that are close in time is large, an image captured by a normal camera has a blur due to the motion even in one image. It is hard to produce a difference of two pixel units. That is, in such a case, even if a motion vector is obtained from an image that is farther apart in time, the encoding efficiency is little improved despite an increase in the amount of calculation for search. Therefore, by restricting the search range of the motion vector in the temporally distant image to the vicinity of the coding area of the image to be coded, the amount of calculation for the search is further reduced and high coding efficiency is obtained. .

Further, in the present invention, a spatio-temporal filter process is performed on a reference image in an area specified by a motion vector between input images to generate prediction signal candidates, and a prediction signal is determined from these candidates. Thus, prediction suitable for an input image input in step 2, ie, an interlaced image, can be performed, and the prediction accuracy is improved. In this case, the motion vector for the field to be subjected to the spatial filter processing is coded with an accuracy of one pixel or less, but the motion vector for the other fields is a motion vector coded with the precision of one pixel or less in the field. By encoding and transmitting the difference between the converted motion vector and the motion vector in the field, information on the finally transmitted motion vector is reduced.

[0032]

Embodiments of the present invention will be described below. FIG. 1 is a block diagram of a moving picture coding apparatus according to the first embodiment of the present invention. 1 includes an encoding unit 14, two field memories 15, 16, a motion vector search circuit 17, an interpolation value generation circuit 19, and a local decoding unit 25.

The image signal 11 to be encoded is encoded by the encoding unit 14 using the reference image signal 12 output from the interpolation value generation circuit 19 and the optimal motion vector 13 output from the motion vector search circuit 17. You. Reference image signal 12
A interpolation value generation circuit is used to calculate an interpolation value generation circuit from an image signal at a position designated by the motion vector candidate 18 output from the motion vector search circuit 17 among the encoded image signals of the past two fields stored in the field memories 15 and 16. 1
9.

The motion vector search circuit 17 comprises a correlation operation circuit 20, a motion vector candidate generation circuit 21, an optimum vector determination circuit 22, and a switch 23. During the motion vector search operation, the motion vector candidate generation circuit 21 generates sequentially. The motion vector candidate 24 is input to the field memories 15 and 16 by the switch 23, and based on this, the correlation operation between the reference image signal 12 and the encoding target image signal 11 generated by the interpolation value generation circuit 19 is performed. This is performed by the circuit 20. The optimal vector determination circuit 22 stores a motion vector that maximizes the correlation between the encoding target image signal 11 and the reference image signal 12, and outputs the optimal motion vector 13 by the switch 23 when the motion vector search operation ends. The optimum motion vector 13 and the interpolation value generation circuit 19
The encoding target image signal 11 is encoded by the encoding unit 14 based on the optimal reference image signal 12 from, and encoded data 26 is output.

The local decoding unit 25 generates a local decoded image signal 27 based on the encoded data 26 output from the encoding unit 14.
Create The local decoded image signal 27 is input to one of the field memories 15 and 16 by a switch 28, and the output thereof is passed through a switch 29 to the interpolation value generating circuit 1.
9 is input. Here, the switches 28 and 29 are used to generate a predetermined reference image for the image to be encoded.
Switching is performed so that the image signal is input to the interpolation value generation circuit 19.

The interpolation value generation circuit 19 comprises an intra-field interpolation circuit 30, a multiplier 31, a multiplier 32 and an adder 33, and a signal generated by the intra-field interpolation circuit 30 from an output signal from the field memory 15; The output signal from the field memory 16 is mixed with the mixing ratio k: 1−k.
To obtain a reference image signal 12.

The motion vector candidate 18 output from the motion vector search circuit 17 is also input to an interpolation value generation circuit 19, and controls a parameter k for determining a mixing ratio for output signals from the field memories 15 and 16. That is, when the vertical component of the motion vector candidate 18 is n + ／ line (n is an integer) in the field, k = 1 is controlled, and the corresponding pixel value stored in the field memory 15 (here, the field memory 15 Described below is an image close to the image to be encoded) is output as the reference image signal 12 as it is.

If the vertical component of the motion vector candidate 18 is n lines in the field, the field memory 1
Using the output signals from 5 and 16, the interpolation value に お け る in FIG.
Is interpolated. At this time, if the absolute value of the vertical component of the motion vector candidate 18 is larger than a certain threshold value, the motion in the reference image is also regarded as large, and the interpolation value 隣接 is adjacent to the interpolation value 内 in the field memory 15 of the same field. Since it is appropriate to mainly use the signal to be interpolated, the parameter k is set large. Conversely, when the absolute value of the vertical component of the motion vector candidate 18 is smaller than the threshold value, the motion in the reference image is also regarded as small, and the interpolation value △ is mainly used as a signal adjacent to the interpolation value 内 in the field memory. Therefore, the parameter k is set small.

Note that, as shown in FIG.
4 and the reference image 41 are adjacent to each other, it is effective to approach the parameter k to 0 and use the signal from the field memory 16 as an interpolation value of the reference image 41.
It is almost limited to the case where the absolute value of the vertical component of the motion vector candidate 18 is 0. On the other hand, when the encoding order as shown in FIG. 3 is adopted, as shown in FIG.
3 and the reference image 51 may be relatively far from each other. In such a case, using the signal from the field memory 16 as the interpolation value of the reference image 51 by setting the parameter k close to 0 is effective even when the absolute value of the vertical component of the motion vector is not 0. come.

FIG. 5 shows a second embodiment of the present invention. The configuration other than the configuration of the interpolation value generation circuit 19 is the same as that of FIG.
Detailed description is omitted. The interpolated value generating circuit 19 is composed of two interpolated value generating circuits 34 and 35, and a switch 36 for selecting the output of the two. Interpolated value generation circuit 3
Reference numerals 4 and 35 denote the interpolation value generating circuit 19 shown in FIG.
Similarly, the intra-field interpolation circuit 30 and the multipliers 31, 3
2 and an adder 33. FIG. 6 shows the relationship between image signals in this embodiment.

The first interpolation value generation circuit 34 generates the reference image 6
When the amount of motion at 1 is large, an effective interpolation value is generated. That is, as shown in FIG. 6A, when the vertical component of the motion vector candidate 18 is n + ／ line (n is an integer) in the field, the corresponding pixel value （(here, In the following description, it is assumed that the content of the field memory 15 is an image close to the encoding target image 63), which is output as a reference image signal (controlled to k = 1).

When the vertical component of the motion vector candidate 18 is n lines in the field (n is an integer), the first interpolation value generation circuit 34 stores the data in the field memory 15 adjacent to the pixel す べ き to be interpolated. An interpolation value に よ っ て is created by averaging two pixel values の of the image signal 61.

The second interpolation value generation circuit 35 generates the reference image 6
When the amount of motion at 1 is small, an effective interpolation value is generated. That is, as shown in FIG. 6B, when the vertical component of the motion vector candidate 18 is n + ／ lines (n is an integer) in the field, the corresponding pixel value ○ stored in the field memory 15 (here, , Field memory 1
5 is assumed to be an image close to the encoding target image 63), is output as it is as a reference image signal (k = 1). For example, when n = 0, the reference image signal for the encoding target image signal 64 has a pixel value of 65
Becomes

When the vertical component of the motion vector candidate 18 is n lines in the field, the second interpolation value generation circuit 35 sets the pixel ◎ to be interpolated to the signal の in the adjacent field memory 16. For example, when n = 0, the interpolation value 68 of the reference image signal with respect to the encoding target image signal 64
Becomes a pixel value 66.

If the vertical component of the motion vector candidate 18 is n / 2 + ／ line in the field, the pixel value ○ from the field memory 15 adjacent to the pixel す べ き to be interpolated and the pixel value Interpolation value △ is created by averaging with ○. For example, if n = 0,
The interpolation value 67 of the reference image signal with respect to the encoding target image signal 64 is an average value of the pixel values 65 and 66.

As described above, the interpolation value generation circuit 35 can generate a reference image signal with an accuracy of 1/4 line unit in a field, and can perform effective motion compensation for an image with little motion and high resolution. I do.

Here, when the absolute value of the vertical component of the motion vector candidate 18 output from the motion vector search circuit 17 is larger than a certain threshold, the switch 36 selects the output of the interpolation value generation circuit 34 and If smaller, interpolation value circuit 3
By selecting the 5 outputs, an appropriate reference image signal is generated for each output. As another method, a motion vector candidate 18 output from the motion vector search circuit 17 is used.
When the absolute value of the vertical component is large, the output of the interpolation value circuit 34 is selected. When the absolute value of the vertical component is small, both outputs of the interpolation value generation circuits 34 and 35 are used as reference image signals. The method of selecting either one is also effective.

Next, a third embodiment of the present invention will be described. FIG. 7 is a block diagram of a moving picture coding apparatus according to the third embodiment of the present invention.

In FIG. 7, image data input to the input terminal 001 is temporarily stored in the input buffer memory 100, and then, in accordance with the order of the image to be coded, for each partial region composed of a plurality of pixels. It is read from the partial image data input buffer memory 100. The partial image data read from the input buffer memory 100 is input to the motion vector detection circuit 200. The motion vector detection circuit 200 obtains a partial image capable of efficiently encoding the input data from the reproduced images encoded in the past, and obtains the data of the area of the partial image and the motion vector (address data) indicating the position of the area. Is output.

The partial image data output from the input buffer memory 100 is also input to the local encoding circuit 300 together with the partial image data and the motion vector output from the motion vector detection circuit 200. The local encoding circuit 300 encodes either partial image data itself output from the input buffer memory 100 or difference data from partial image data specified by a motion vector. Here, the encoded data of the difference from the area specified by the motion vector includes data obtained by performing variable length encoding on the motion vector.

The data encoded by the local encoding circuit 300 is input to a local decoding circuit 400 together with the partial image data output from the motion vector detection circuit 200 and decoded, whereby a reproduced image is obtained. When the encoded data is encoded using a motion vector, a decoded image is added to the partial image data output from the motion vector detection circuit 200 to obtain a reproduced image.
The reproduced image data is input to the motion vector detection circuit 200, and is temporarily stored for encoding the next input image data.

Next, a specific operation example of the motion vector detecting circuit 200, the local encoding circuit 300, and the local decoding circuit 400 will be described.

First, in the motion vector detection circuit 200, the data input from the input buffer memory 100 is an image memory (any one of 211 to 214) in which image data which is no longer necessary for searching for a motion vector is stored.
Are sequentially written by the write control circuit 222. The encoded image data stored in the image memories (211 to 214) in this manner is sequentially read out by the read control circuit 221 and the data switching circuit 231 from an image temporally close to the encoded image for each area. , Difference circuit 241
The difference from the input data is calculated for each area. The evaluation circuit 242 sequentially compares the magnitudes of the sums of the differences for each area, controls the direction in which the readout control circuit 221 searches the image memory, and inputs the partial image input from the previously detected partial image. Each time a partial area of an encoded image having a smaller difference is detected, an address indicating the area of the partial image is stored in the vector register 243, and the partial area of the encoded image closest to the input partial image is stored. Ask for. The address data indicating the partial area of the encoded image having the smallest difference from the input partial image and stored in the vector register 243 in this manner is input to the readout control circuit 233 and the switching circuit 232, and the encoding is performed. The reproduced image of the encoded data corresponding to the partial region of the completed image is read from the reproduced image memories (215 to 218), and is input to the local encoding circuit 300 together with the address data.

On the other hand, in local encoding circuit 300,
In this embodiment, DCT (Discrete Cosine Transform), which is one of orthogonal transform, quantization and variable length coding are used as a method of coding a motion compensation error. Local coding circuit 300
First, the partial image data output from the input buffer memory 100 is input to the difference circuit 311, and the partial image data is compared with the partial image data output from the motion vector detection circuit 200 and reproduced from the encoded data. A difference is calculated. The switching circuit 312 includes the difference circuit 31
The differential image data input from the input buffer memory 1 and the partial image data input from the input buffer memory 100 are sequentially switched and output by a control signal input to a terminal 002.

The DCT circuit 320 sequentially converts the frequency of the partial image data and the difference image data sequentially output from the switching circuit 312 and outputs the result. The quantization circuit 330 calculates D
The frequency-converted data output from the CT circuit 320 is quantized by a preset quantization width and output. The entropy encoder 340 encodes the data quantized by the quantization circuit 330 together with the quantization width information and the identification code of the partial image data and the difference data, and further encodes the difference image data by a vector. The motion vector for the partial image output from the register 243 is also variable-length coded using a Huffman code or the like corresponding to each occurrence probability and output. If one Huffman code is formed by combining the identification code and the motion vector, efficient encoding is possible. In this encoding, image data obtained by reproducing encoded data in an area specified by a predetermined rule, or input image data whose difference from fixed data is equal to or less than a set value, is not changed. If the number of consecutive partial images is variable-length coded, the coding efficiency is further improved.

The code amount evaluation circuit 351 calculates the code amount between the case where the difference between the partial image to be coded and the area specified by the motion vector is coded and the case where the input data is coded by DCT as it is. The comparison is performed, and the encoded data with the higher encoding efficiency is output to the output buffer 360 and the local decoding circuit 400.

The output buffer 360 temporarily stores the encoded data in order to adjust the output data rate, and also uses the quantization width used by the quantization circuit 330 and the encoding used by the entropy encoder 340. Control the table.

In local decoding circuit 400, the partial image data output from motion vector detecting circuit 200 is temporarily stored in data memory 441, and the encoded data output from code amount evaluating circuit 351 is stored in a variable length decoder. A motion vector including an identification code and quantized data before encoding are input to 410 and decoded. The decoded quantized data is input to the inverse quantization circuit 420, converted into a representative value having a dynamic range before quantization (inverse quantization), and then input to the addition circuit 450. The data inversely quantized by the inverse quantization circuit 420 is output to the inverse DCT circuit 43.
0, and the partial image data or the difference image data is reproduced. When the reproduced data output from the inverse DCT circuit 430 is differential image data, the gate circuit 442 allows the partial image data output from the data memory 441 to pass, based on the identification code decoded by the variable length decoder 410, If the image is not a difference image, the output data is set to zero and output to the adder circuit 450, respectively.

If the image data subjected to the inverse DCT is obtained by encoding the difference image, the image data is added to the partial image data output from the motion vector detection circuit 200. Circuit 2
A reproduced image is obtained from the adder circuit 450 without using the partial image data output from 00. The reproduced image data is input to the motion vector detection circuit 200, and is temporarily stored for encoding the next input image data.

FIG. 8 shows a fourth embodiment of the present invention.
In that a data memory 460 for storing quantized data is used instead of the variable length decoding circuit 410 included in the encoding circuit 400 of FIG. In this case, data obtained by DCT and quantization of difference image data between the partial image data output from the input buffer memory 100 and the reproduced image specified by the motion vector is stored in the data memory 4.
60 is temporarily stored. Then, the code amount evaluation circuit 352
, And image data corresponding to the encoded data output to the output buffer 360 is output to the inverse quantization circuit 420. When this image data is difference image data, a reproduced image is obtained by adding the partial image data output from the data memory 441 and the addition circuit 450 in the same manner as in the example of FIG.

This embodiment has an advantage that the processing time is shorter than that in FIG. 1 because the operation for decoding is unnecessary.

FIG. 9 shows a fifth embodiment of the present invention, in which a part of an encoding circuit 400 is used for a motion vector detecting circuit. Also in this embodiment, the motion vector detecting circuit 200 switches the partial image data from the already encoded and reproduced image according to the read control circuit 224.
At the same time, a motion vector (address data) indicating the position of the area is output.

In FIG. 9, the image data input to the input terminal 001 is temporarily stored in the input buffer memory 100 as in FIGS. 7 and 8, and is composed of a plurality of pixels according to the order of the image to be encoded. Read from the input buffer memory 100 for each region to be
Input to 0. The difference between the partial image data output from the input buffer memory 100 and the output data from the motion vector detection circuit 200 input via the gate circuit 313 by the control signal input to the input terminal 002 is not first obtained. 7 and 8, DCT, quantization, and variable-length coding are performed. The quantized data is stored in the data memory 460, and the coded data and the code amount are stored in the code amount evaluator 353. Next, the input terminal 002
Is changed, the difference from the partial image data output by the motion vector detection circuit 200 is calculated, and DCT, quantization, and variable length coding are performed as in FIGS. 7 and 8. .

The code amount evaluator 353 evaluates the encoded data from the entropy encoder 340 and controls the read control circuit 224 according to the evaluation result, so that every time an area where the code amount can be further reduced is detected. In addition to storing the encoded data, the motion vector detecting circuit 200
Is stored in the data memory 441, and the data obtained by DCT of the difference image and quantized is stored in the data memory 460. In this way, the encoded data finally having the smallest code amount is output to the output buffer 360. The quantized data corresponding to the encoded data is supplied to the local decoding circuit 400 as shown in FIGS.
Is reproduced in the same way as. The reproduction data is input to the motion vector detection circuit 200, and is temporarily stored for encoding the next input image data.

In the embodiments of FIGS. 7 and 8, a more accurate motion vector is obtained as the optimum motion vector, whereas in the embodiment of FIG. 9, a motion vector that maximizes the coding efficiency is obtained.

FIG. 10 shows an example of a circuit for reproducing encoded data in the embodiment shown in FIGS. The variable-length decoder 510 decodes the coded data input from the input terminal 004, and includes quantization width information, a motion vector (including a difference image data from a partial image specified by the motion vector or an identification code indicating whether or not the difference image data is a motion vector) ), And reproduce the quantized data. The quantized data is supplied to an inverse quantization circuit 52.
0, reproduced through the inverse DCT circuit 530. If the reproduced data is differential image data from the partial image specified by the motion vector, the read control circuit 521 reads the partial image data from the reproduced image memories 611 to 614 and outputs the read partial image data from the motion vector detection circuit 600. Input to the decoding circuit 500. This partial image data is added to the difference image data by the addition circuit 550 via the gate circuit 540 to become reproduced image data. This reproduced image data is input to the motion vector detection circuit 600 for the reproduction of the next input coded data, and is temporarily stored. The reproduced image data is also input to the output buffer 560, and is returned to the original image order and output.

Next, the motion vector search operation performed in the evaluation circuit 242 and the read control circuit 221 in FIGS. 7 and 8 and the code amount evaluator 352 and the read control circuit 224 in FIG. 9 will be described.

FIGS. 11 to 16 show examples of motion vector search according to the present invention. These FIGS. 11 to 16
, S1, s2, ..., s6 are one frame or one
.., 12 mean field images
0 means one or more pixels in the horizontal or vertical direction.

In the example of the motion vector search shown in FIG. 11, the entire encoded image s3 (10%) which is temporally close to the partial image (eg, s4-104) to be encoded
1 to 120), and the optimal motion vector obtained by the search, that is, a coding efficiency is high or a more accurate motion vector (for example, s3-107) has been encoded into the encoding target image based on the motion vector. The search range (area) in the encoded image s2 that is temporally separated from the image s3 is limited to, for example, (105 to 115).

In the search for the encoded image s2 separated in time, the optimal motion vector (for example, s02-109 or s03-107) including the previously obtained optimal motion vector (for example, s03-107) is included. ).

Then, the difference between the image in the area specified by the obtained optimal motion vector (for example, s2-107) and the partial image to be coded (for example, s4-104), that is, the motion compensation error is obtained. Encode the optimal motion vector and motion compensation error.

The example of the motion vector search shown in FIGS. 12 to 14 is an image (for example, s4) that is temporally close to the encoding target image s4.
This is a preferred example in the case where 3) is encoded using a motion vector between a temporally separated image (for example, s1). In this case, since it is possible to predict a motion vector between the encoding target partial image s4-104 and the temporally close image s3, the temporally close image s3 is obtained.
Is limited to, for example, (103-112). If a motion vector (for example, s3-107) having a high coding efficiency with respect to a temporally close image is obtained, a partial region (for example, s3-107) of a temporally close image specified by the motion vector is obtained. 107), using a motion vector (for example, s1-110) between the temporally distant image (for example, s1) used for the partial image s4-104 of the encoding target image s4. The area that is a candidate for a motion vector between the image and the image (for example, s2) can be limited to a narrower range, for example, (107 to 110).

In the example of the motion vector search shown in FIG. 15, the motion between the images is very large, and there is no motion vector that is a candidate even in the image s3 that is close in time. This is an example of a case where the obtained motion vector (for example, s3-116) is very large, and it can be predicted that there is no motion vector with good coding efficiency in the image s2 separated in time.

In such a case, the image s2 separated in time
For, the search for the motion vector may not be performed, or the search range may be further limited to, for example, (116 to 120). This makes it possible to obtain a motion vector with good coding efficiency even if the amount of computation required for searching for a motion vector is small.

In the example of the motion vector search shown in FIG. 16, an image s3 temporally close to the partial image s4-104 to be encoded is displayed.
This is an example of a case where the motion vector (for example, s3-107) obtained in step (1) is large to some extent, or a case where an appropriate motion vector cannot be obtained. In these cases, the search for the motion vector in the temporally distant image s2 is not performed, and the search in the temporally distant image s2 is performed only when the motion vector between the temporally distant image s3 is small. The range is set near the same position as the coding area of the coding target image s4 (for example, 103
To 105) to search for a motion vector.

If the motion between images close in time is large, "blur" due to the motion exists in one image captured by a normal camera. It is hard to produce a difference in pixel units. That is, in such a case, even if a motion vector is obtained from an image that is farther apart in time, the encoding efficiency is little improved despite an increase in the amount of calculation for search. According to the example of FIG. 16, in such a case, since the motion vector of a temporally distant image is not encoded in an image with a large motion, the types of encoding are reduced, and the encoding efficiency is further increased.

The encoding method according to the present invention may be any of intra-frame and intra-field encoding, or inter-frame and inter-field differential encoding, and may be combined with another encoding method. In addition, the motion vector search method according to the present invention is also applicable to a case where a motion vector is calculated for an image between them by using an encoded image previously encoded at an arbitrary number of images as a reference image. The search can be performed from before and after the target image.

Next, another embodiment of the present invention will be described. FIG. 17 is a block diagram of a video encoding device according to the sixth embodiment of the present invention. This embodiment is an example in which the present invention is applied to an encoding system such as SM3 which is a simulation model of MPEG1, and the encoding algorithm basically employs motion compensation + DCT (discrete cosine transform). . However, the input image format is, for example, CCIR Rec. This is an interlace format such as the image format defined in 601/525.
This image format is shown in FIG. In FIG. 18, Y is a luminance signal, Cr and Cb are chrominance signals, and the number of pixels per field of each signal is as shown in the figure.

The coding unit is hierarchically structured, similarly to SM3. That is, as shown in FIG. 19, the layers are hierarchized in the order of blocks, macroblocks, slices, and pictures, and a group of pictures and a sequence (not shown in FIG. 19). The block is 8x
DCT is performed in units of blocks. The macro block is composed of two Y blocks and one each of a Cr block and a Cb block, for a total of four blocks. The motion compensation and the selection of each encoding mode are performed in units of this macro block.

A group of pictures (GOP) is configured as follows. Pictures are largely divided into I, P, and B pictures according to the type of mode for each macroblock that is allowed as a prediction mode, similarly to SM3. As modes, there are four modes of intra-field prediction (Intra), forward prediction (Inter: including motion compensation), backward prediction, and bidirectional interpolation prediction (details will be described later). Of these, picture types are classified into three types, I, P, and B, as shown in Table 1, depending on which prediction mode can be used.

[0081]

[Table 1] In the present embodiment, unlike SM3, an interlaced format is adopted as a coded picture format, and since the prediction method and the like differ depending on the position in the GOP even for the same type of picture, the pictures are further classified. FIG. 20 shows the structure of a GOP and from which picture each picture is predicted. As shown in the figure,
GOP starts from B ₀ picture preceding random access and special playback is periodically provided as entry points for I-pictures, are defined by a set of pictures that end with P ₂ preceding the next I picture. I-pictures appear only in even fields.

Table 2 and FIG. 21 show how each picture is predicted from which picture.

[0083]

[Table 2] There are two types of prediction methods: inter-field prediction in which prediction is performed only from even fields, and inter-field / inter-frame adaptive prediction in which prediction is performed adaptively from even / odd fields. In FIG. 21, an arrow indicates that the prediction is performed for inter-field prediction, and a symbol in which two lines are integrated into one arrow indicates that the inter-field / inter-frame adaptive prediction is performed. The encoding order of each picture in one GOP is as shown in FIG.

Based on the above points, the moving picture coding apparatus shown in FIG. 17 will be described. In FIG. 17, an input terminal 70
To 0, an interlaced moving image signal is input.
The input moving image signal is stored in the field memory 701 for eight consecutive fields. Frame memory 701
Are supplied to a first motion vector detecting circuit 710, and a motion vector is detected with one-pixel accuracy by a telescopic search (described later) using a primary moving image. This is the first stage of the motion compensation.

Next, as preprocessing in the second stage of the motion compensation, the motion vector detection circuit 7 uses the local decoded signal stored in the field memory 708 in the local decoding loop.
In step 10, the motion vector obtained using the original image is refined by performing a full search in a range of ± 1 around the motion vector.

Next, as a second stage of motion compensation, a field memory 708, an inter-field / inter-frame adaptive prediction circuit 709, and a second motion vector detection circuit 711
In 動 き, a motion vector is detected with half-pixel accuracy using a local decoded signal, and a prediction signal is generated by a prediction circuit 709. This prediction signal is input to the subtractor 702, and a difference from the moving image signal from the field memory 701 is obtained, and this difference is output as a prediction error signal.

The prediction error signal is subjected to discrete cosine transform by the DCT circuit 703 to obtain DCT coefficient data. The DCT coefficient data is quantized by a quantization circuit 704 and adaptively scanned.
The signal is input to the multiplexer 714 via 10. The quantized DCT coefficient data is locally decoded via an inverse quantization circuit 705 and an inverse DCT circuit 706, and only I and P pictures are written to a field memory 708. The field memory 708 is prepared for four fields required for adaptive prediction.

The multiplexer 714 multiplexes the DCT coefficient data, the motion vector from the second motion vector detection circuit 711 described later, the macroblock type from the coding control unit 717, and additional information such as the step size. . The information multiplexed by the multiplexer 714 has a constant transmission rate via a buffer 715 and is output to a storage system (recording medium) such as a VTR.

In the encoding control unit 717, the buffer amount in the buffer 715 and the activity calculation circuit 71
The quantization step size in the quantization circuit 704 is calculated using the intra-macro-block activity (I picture) calculated in step 6 or the immediately-before-quantized signal of the same mode before the quantization (P, B-picture). Controlling.

Next, the configuration of a moving picture decoding apparatus corresponding to the moving picture coding apparatus of FIG. 17 will be described with reference to FIG. A signal read from the storage system is input to the input terminal 800 and temporarily stored in the buffer 801. Buffer 801
Are input to the demultiplexer / variable-length decoding circuit 801 and the multiplexer 71 shown in FIG.
The additional information such as the DCT coefficient data, the motion vector, and the step size multiplexed in step 4 are separated and decoded.

That is, the DCT coefficient data is subjected to two-dimensional variable length decoding and scan conversion, and then to the inverse quantization circuit 8 as in the local decoding loop in the moving picture coding apparatus of FIG.
03, and locally decoded via the inverse DCT circuit 804, the adder 805 and the field memory 8 for 4 fields.
06 and input to the adaptive prediction circuit 807. On the other hand, since the motion vector is sent in the form of a difference as described later, after the variable length decoding, the motion vector is returned to the original form and supplied to the adaptive prediction circuit 807. The adaptive prediction circuit 807 generates a prediction signal, and the prediction signal and the local decoded signal from the inverse DCT circuit 804 are added by the adder 805, so that
The original moving image signal is extracted via the field memory 806. The output of the field memory 806 is supplied to an image display unit (not shown).

However, the display order of the picture is different from the decoding order and the display order, and since the B picture is not used for prediction, it need not be stored in the field memory.
The output of the I and P pictures is output from the field memory 806, and the output of the B picture is decoded and output as it is.

Next, the operation of motion compensation and motion vector detection in this embodiment will be described in detail. Motion compensation is
As described above, the processing is performed for each macroblock. The motion vector detection is performed by the first motion vector detection circuit 710.
For each point obtained by performing motion vector detection by ordinary block matching with pixel accuracy, and performing adaptive spatio-temporal interpolation on pixel positions with 1/2 pixel accuracy around the reference image specified by the 1-pixel accuracy motion vector , Motion vector detection in the second motion vector detection circuit 711 by searching for a motion vector. The specific method of spatiotemporal interpolation will be described later.

The motion vector detection in the first motion vector detection circuit 710 is a process of finding an optimal motion vector by performing a full search for each field with one pixel accuracy using each field image of the input moving image. A search between distant fields includes a telescopic search (SM
3). However, in a field image, a plurality of paths can be considered as search paths between fields. In this embodiment, the search path is determined according to the following rules. 1) Search between in-phase fields The search is performed using only in-phase fields. 2) Search between opposite phase fields Search is performed using the same phase field as much as possible.

Although it is necessary to include a search between the opposite phase fields only once in the path, in the present embodiment, as shown in FIG. 24, the order of the telescopic search is such that fields having different phases are searched first. I have to. For example I
→ when determining the motion vector of the P _{2 is, I → P o → B 1} →
I searched in the order of _{B 3 → P 2, I →} B 0 → B 2 → P 1 → P
No route like ₂ is taken. In FIG. 24, a range indicated by hatching is a search range. This telescopic search is performed separately for the forward direction and the reverse direction. The search range between adjacent fields is ± 15 horizontal pixels and ± 7 vertical pixels.

The motion vector detection in the second motion vector detection circuit 711 is performed using an image stored in the field memory 708 in the local decoding loop as a reference image. First, as preprocessing of the motion vector detection, the first
By performing a full search within a range of ± 1 pixel around the reference image specified by the motion vector obtained by the motion vector detection circuit 710, the motion vector is refined. Second
The main process of detecting a motion vector by the motion vector detection circuit 711 is a plurality of prediction signals generated by a method described later at a position of 1/2 pixel accuracy around a reference image specified by a motion vector obtained by refinement. The evaluation is performed by evaluating and comparing all prediction error powers of the candidates, thereby selecting an optimum prediction signal candidate.

Note that motion vector detection is not performed on the color signal, and motion compensation is performed based on the motion vector obtained from the luminance signal.

Next, the processing of the inter-field / inter-frame adaptive prediction circuit 709 in FIG. 17 will be described. FIG.
FIG. 14 is a block diagram showing a partial configuration of an inter-field / inter-frame adaptive prediction circuit 709.

As described above, the second stage of the motion compensation is a search for a range of 1/2 pixel accuracy around the reference image specified by the motion vector obtained by the first motion vector detection circuit 710. FIG. 25B shows this state. This inter-field / inter-frame adaptive prediction is represented by P ₁ , P ₂ , B
Is performed on the current picture to be coded (coded field) using, for example, a pair # 1 and # 2 of each preceding even and odd field as a reference picture (reference field).

In the first stage of the motion compensation shown in FIG. 25A, the optimal point (indicated by ●) in each of the reference fields # 1 and # 2 is determined by the motion vector detecting circuit 710, respectively. Suppose that it was obtained by FIG.
In the second stage of the motion compensation shown in (b), の around the two optimal points specified by the motion vectors V1 and V2.
The inter-field / inter-frame adaptive prediction circuit 709 applies a spatio-temporal filter to the reference image in the pixel-accuracy region to obtain a plurality of prediction signal candidates. ,
The motion vector detection circuit 711 searches for an optimal motion vector. In this case, there are two modes for generating a prediction signal candidate: a field interpolation mode and a frame interpolation mode.

First, in the field interpolation mode, a prediction signal candidate is generated only by a spatial filter among spatio-temporal filters. That is, the pixel values of the pixels A to C in FIG. 25B are created by A = (O + D) / 2 B = (O + D + E + F) / 4 C = (O + F) 2 In this field interpolation mode, there are a total of 18 search signal candidate search points, 9 in each of the even and odd fields.

On the other hand, in the frame interpolation mode, a prediction signal candidate is created by performing a spatio-temporal filter process on a signal subjected to motion compensation with one-pixel accuracy in each field. For example, the pixel values of the pixels A to C in FIG. 25B are created by A = G / 2 + (O + D) / 4 B = G / 2 + (O + D + E + F) / 8 C = G / 2 + (O + F)） You. In this case, the field that provides the data at the pixel position with 1/2 pixel precision is referred to as a reference field. Also in this frame interpolation mode, there are nine prediction signal candidates in each of the even and odd fields, but since the prediction signals at the position of the pixel O match, there are a total of 17 search signal candidate search points.

In the second stage of motion compensation, the sum of
By searching for five prediction signal candidates, a prediction signal candidate having the smallest prediction error is determined as a prediction signal. However, when the directions of the motion vectors pointing to both fields are largely shifted, the frame interpolation mode is not selected (details will be described later).

The information indicating whether the field interpolation mode or the frame interpolation mode is selected and the information indicating which field is selected as the reference field used in the field interpolation mode or the reference field in the frame interpolation mode are 1-bit. Transmitted by flag. Note that the prediction by the field immediately before P ₀ and P ₂ is performed only by a mode in which the field interpolation mode is adapted to a single field. In this case, the mode selection flag is not transmitted.

While the above description has been made with reference to a P picture, the same applies to a B picture. However, in the case of a B picture, only a field having the same phase as the coding field is selected as a search field in the field interpolation mode and a reference field in the frame interpolation mode. Of course, in this case, a flag indicating which field is selected is not transmitted.

Next, how the above-described principle is actually realized on hardware will be described with reference to FIG. Cache memories 901a, 901b and 902
In a and 902b, a portion of the image signal from the field memory 708 corresponding to four fields of the local decoding loop in FIG. 17 designated by the motion vector obtained in the pre-processing of the second motion vector detection is stored. You. Switching circuit 900
Distributes the outputs of the cache memories 901a and 901b to the time filter 903 and the spatial filter 905 according to the control signal from the second motion vector detection circuit 711,
The outputs of the cache memories 902a and 902b are distributed to a time filter 904 and a spatial filter 906.

In accordance with a control signal from the second motion vector detection circuit 711 indicating which of the field interpolation mode and the frame interpolation mode is to be selected, the selector 907 determines whether the signal passed through only the spatial filter 905, the spatial filter 905 and the time One of the signals passing through both of the filters 903 is selected. The same applies to the selector 908. For example, from the switching circuit 900 to the time filter 903
If the signal of the pixel G in FIG. 25B is input to the pixel and the signal of the pixel B in FIG. 25B is output from the spatial filter 905, the signals of the pixels G and B are output to the output of the time filter 903. An averaged signal, that is, a signal subjected to spatio-temporal filtering is obtained. Therefore, the output of the spatial filter 905 may be selected in the field interpolation mode, and the output of the temporal filter 903 may be selected in the frame interpolation mode.

Outputs of the selectors 907 and 908 are directly input to the selector 911, and are added by the adder 909 and halved by the divider 910 and then input to the selector 911. The selector 911 includes a signal predicted using these three inputs, that is, a signal predicted using the reference fields # 1 and # 2 from the selector 907, a signal predicted using the reference fields # 3 and # 4 from the selector 908, and the divider 91.
Signals predicted using reference fields # 1 to # 4 starting from 0 are selected and output to the second motion vector detection circuit 711 as predicted signal candidates. Motion vector detection circuit 71
In step 1, as described above, a prediction signal candidate having the smallest prediction error is determined as a prediction signal from among the 35 prediction signal candidates, and information indicating this is returned to. Thereby, the inter-field / inter-frame adaptive prediction circuit 709 is provided.
Outputs the prediction signal indicated by the motion vector detection circuit 711 to the subtractor and adder 707.

Next, the second motion vector detection circuit 711
In what form the motion vector from is encoded and transmitted will be described. First, a motion vector in the field interpolation mode or a motion vector on the reference field side in the frame interpolation mode is encoded as a reference vector by the variable length encoding circuit 713 and sent to the multiplexer 714. This motion vector (reference vector) has 1/2 pixel accuracy. If the frame interpolation mode is selected, the motion vector of the field other than the reference field and the motion vector of the reference field (reference vector)
Is converted to the corresponding field and the difference is similarly encoded and transmitted. That is, the difference between the extension line of the motion vector in the reference field and the point closest to the intersection of the non-reference field and the motion vector is 1
It shall be sent with pixel accuracy. Except when the directions of the two motion vectors are close to each other, the frame interpolation mode is considered to be ineffective. No, in other words, the frame interpolation mode is used only when the magnitude of the difference is equal to or smaller than a minute value, in this example, equal to or smaller than 1.

[0110] Figure 26 shows a specific example of how to send this motion vector, the first reference field # 1 and code
Between the encoding target field (the encoding field in the figure)
The first motion vector with half-pixel accuracy (the upper arrow in the figure)
), The second reference field # 2 and the encoding target
The second motion vector with one pixel accuracy between the
(Indicated by the middle and lower arrows), and the second motion vector
Motion vectors (indicated by ○ and ●)
Vector whose direction is closest to the first motion vector
The first motion vector and the second reference field
The vector of the point ● closest to the point △ where the de # 2 intersects
The difference from the motion vector (the vertical arrow in the figure)
A difference vector with a corresponding one-pixel accuracy is obtained, and the first motion
And information on the difference vector is encoded and sent. In the example of FIG. 26, the difference is −1. Depending on this way, without lowering the prediction performance, it is possible to save the information amount of the motion vector.

Next, the motion compensation of the color signal will be described.

As shown in FIG. 19, in one macroblock, the luminance signal Y and the chrominance signals Cr and Cb have the same number of pixels in the vertical direction, but the number of pixels of the chrominance signal is halved in the horizontal direction. ing. Therefore, when a motion vector obtained from a luminance signal is applied to a color signal, the horizontal component of the motion vector is reduced to half. The division by ２ is performed by rounding an integer or less toward zero. This is the same in the field interpolation mode and the frame interpolation mode.

FIGS. 27 (a) and 27 (b) show specific examples in which a motion vector of a color signal is obtained from a motion vector of a luminance signal in the field interpolation mode and the frame interpolation mode. In FIG. 27, circles indicated by dotted lines indicate pixel positions where no color signal exists. The pixel position obtained in the first step is represented by a central circle, and the point obtained in the second step is represented by x. Further, in this example, in any case, the horizontal coordinate of the origin is larger than the horizontal coordinate of the center circle point. In the case of the field interpolation mode shown in FIG.
When the motion vector of the luminance signal is obtained at the position of, the 信号 pixel precision of the color signal is rounded in the direction A to create an interpolation pixel △. This interpolated pixel 作成 is created by Δ = 1/4 (A + B + C + D). Similarly, for the reference field # 1 in the frame interpolation mode of FIG. 27B, the interpolation pixel △ in the reference field # 1 is created by Δ = 1/4 (E + F + G + H).

On the other hand, as for the interpolated pixel in the reference field # 2, a half pixel precision of the color signal is rounded in the direction A, and the pixel at the position I is used. After all, in the case of the example of FIG. 27, the value of the interpolation pixel △ is obtained by I × 1/2 + (E + F + G + H) × 1/8.

[0115]

As described above, according to the present invention, 2
Efficient and accurate prediction suitable for interlaced images can be performed using two reference fields effectively, and a more accurate predicted signal can be obtained, especially in the moving region of interlaced images. In addition, the increase of the motion vector information accompanying this can be minimized by employing the difference vector.

[0116]

[0117]

[Brief description of the drawings]

FIG. 1 is a block diagram showing a first embodiment.

FIG. 2 is a diagram showing a relationship between image signals of respective screens in the related art.

FIG. 3 is a diagram showing an example in which three adjacent fields are not already encoded;

FIG. 4 is a diagram showing a relationship between image signals of respective screens in the first embodiment.

FIG. 5 is a block diagram showing a second embodiment.

FIG. 6 is a diagram illustrating a relationship between image signals of respective screens according to the second embodiment.

FIG. 7 is a block diagram showing a third embodiment.

FIG. 8 is a block diagram showing a fourth embodiment.

FIG. 9 is a block diagram showing a fifth embodiment.

FIG. 10 is a block diagram of a video decoding device.

FIG. 11 is a diagram showing an example of a motion vector search according to the present invention.

FIG. 12 is a diagram showing an example of a motion vector search according to the present invention.

FIG. 13 is a diagram showing an example of a motion vector search according to the present invention.

FIG. 14 is a diagram showing an example of a motion vector search according to the present invention.

FIG. 15 is a diagram showing an example of a motion vector search according to the present invention.

FIG. 16 is a diagram showing an example of a motion vector search according to the present invention.

FIG. 17 is a block diagram of a video encoding device according to a sixth embodiment of the present invention.

FIG. 18 is a diagram showing an input image format in the embodiment.

FIG. 19 is a diagram showing a hierarchical structure of a coding unit in the embodiment.

FIG. 20 is a diagram showing a configuration of a group of pictures and an encoding order in the embodiment.

FIG. 21 is an exemplary view for explaining a prediction method of each picture in the embodiment.

FIG. 22 is a block diagram of a video decoding device corresponding to the video encoding device of FIG. 17;

23 is a block diagram of an inter-field / inter-frame adaptive prediction circuit in FIG. 17;

FIG. 24 is a diagram showing the order of telescopic search in the embodiment.

FIG. 25 is a view for explaining an inter-field / inter-frame adaptive prediction process in the embodiment.

FIG. 26 is a diagram showing an example of how to send a motion vector in the embodiment.

FIG. 27 is a diagram showing a specific example of obtaining a motion vector of a color signal from a motion vector in the embodiment.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 14 ... Encoding part 17 ... Motion vector search circuit 19 ... Interpolation value generation circuit 20 ... Correlation operation circuit 21 ... Motion vector candidate generation circuit 22 ... Optimal vector judgment circuit 200, 600 ...
Motion vector detection circuit 221, 223, 224, 621 ... image data read control circuit 242, 351, 352, 353 ... evaluation circuit 300 ... local coding circuit 400 ... local decoding circuit 101 ... field memory 103 ... DCT circuit 104 ... quantization Circuit 105 ... Inverse quantization circuit 106 ... Inverse DCT circuit 108 ... Field memory 109 ... Adaptive prediction circuit 110 ... Motion vector detection circuit 111 ... Motion vector detection circuit 112 ... Variable length coding circuit 113 ... Variable length coding circuit 114 ... Mux 115 buffer 116 activity calculation circuit 117 code controller

──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Hideyuki Ueno 1st Toshiba-cho, Komukai-ku, Kawasaki-shi, Kanagawa Prefecture Inside the Toshiba Research Institute, Inc. Address: Toshiba Research Institute Co., Ltd. (72) Inventor: Tomio Yamakage 1, Komukai Toshiba Town, Saiwai-ku, Kawasaki-shi, Kanagawa Prefecture Toshiba Research Institute Co., Ltd. (56) References JP-A-3-226193 (JP, A) Hei 4-268892 (JP, A) Special Table Hei 6-504169 (JP, A) Proceedings of the 1991 IEICE Spring Conference, 7-64 (1991-3) (58) Fields surveyed (Int. Cl. ⁶ , DB name) H04N 7/24-7/68

Claims (84)

(57) [Claims] 1. A first and a second reference field designated by a first and a second motion vector, respectively, between an image of a continuous first and second reference field and an image of a field to be encoded. A mode for generating a prediction signal for an image of the encoding target field from a reference signal on the image, and encoding error information of the prediction signal for an image signal of the image of the encoding target field; Using the mode only when the magnitude of the difference between the vector closest to the first motion vector and the second motion vector among a plurality of motion vectors that can be the motion vector is a minute value or less. A video encoding device characterized by the above-mentioned. 2. The method according to claim 1, wherein the first and second reference vectors are respectively designated by first and second motion vectors between the successive first and second reference field images and the encoding target field image. A mode in which a prediction signal for the image of the encoding target field is generated from a reference signal on the image, and error information of the prediction signal for the image signal of the image of the encoding target field is encoded. A mode used only in a range in which the magnitude of the difference between the vector closest to the first motion vector and the second motion vector among a plurality of motion vectors that can be the motion vector A moving picture coding apparatus comprising: 3. The first and second reference fields designated by first and second motion vectors, respectively, between a picture of a continuous first and second reference field and a picture of an encoding target field. From the reference signal on the image, generate a prediction signal for the image of the encoding target field, error information of the prediction signal with respect to the image signal of the image of the encoding target field, information of the first motion vector, and A mode for encoding information of a difference vector corresponding to a difference between a vector having a direction closest to the first motion vector and a second motion vector among a plurality of motion vectors that can be a second motion vector. A moving image encoding apparatus, wherein the mode is used only when the magnitude of the difference is equal to or smaller than a minute value. 4. The first and second reference fields designated by first and second motion vectors, respectively, between an image of a continuous first and second reference field and an image of a field to be encoded. From the reference signal on the image, generate a prediction signal for the image of the encoding target field, error information of the prediction signal with respect to the image signal of the image of the encoding target field, information of the first motion vector, and In a mode for encoding information of a difference vector corresponding to a difference between a vector whose direction is closest to the first motion vector and the second motion vector among a plurality of motion vectors that can be a second motion vector, A vector having a direction closest to the first motion vector among a plurality of motion vectors that can be the second motion vector; Video encoding apparatus characterized by having a mode in which the magnitude of the difference between the vector is used only in the range of small values. 5. A means for generating first and second motion vectors between successive images of a first and second reference field and an image of a field to be encoded, and said first and second motions. Means for generating a prediction signal for the image of the encoding target field from a reference signal on the image of the first and second reference fields respectively specified by a vector; Means for generating error information of a prediction signal; and a magnitude of a difference between a vector closest in direction to the first motion vector and a second motion vector among a plurality of motion vectors that can be the second motion vector. Encoding means for encoding the error information only when is smaller than a minute value. 6. A means for generating first and second motion vectors between successive images of a first and second reference field and an image of a current field to be coded, and said first and second motions. Means for generating a prediction signal for the image of the encoding target field from a reference signal on the image of the first and second reference fields respectively specified by a vector; Means for generating error information of a prediction signal; and a difference between the vector closest to the first motion vector and the second motion vector among a plurality of motion vectors that can be the second motion vector. Means for generating a difference vector to be calculated, and the error information, the information of the first motion vector, and the information of the difference vector only when the magnitude of the difference is smaller than And a coding means for coding the information respectively. 7. A means for determining a first reference signal on the image of the reference field further specified in the first motion vector between the image of the image and the encoding target field of a first reference field, A motion vector between an image of a second reference field contiguous to the first reference field and an image of the current field, wherein an image signal on the second reference field and a current field before <br/> Symbol second motion vector magnitude of the difference is that Do the following minute value to the nearest vector direction in the first motion vector exist between the images
Means for determining a reference signal on the image of the second reference field specified by the second motion vector only when the reference signal is present; and determining the code from the reference signal on the image of the first and second reference fields. Moving image encoding means for generating a prediction signal for an image of an encoding target field; and encoding means for encoding error information of the prediction signal for the image signal of the image of the encoding target field. apparatus. Means for determining more reference signals on the image of the first reference field specified in the first motion vector between 8. An image of the first image and the encoding target field of the reference field, A motion vector between an image of a second reference field contiguous to the first reference field and an image of the current field, wherein an image signal on the second reference field and a current field before <br/> Symbol second motion vector magnitude of the difference is that Do the following minute value to the nearest vector direction in the first motion vector exist between the images
Means for determining a reference signal on the image of the second reference field specified by the second motion vector only when the reference signal is present; and determining the code from the reference signal on the image of the first and second reference fields. Means for generating a prediction signal for the image of the encoding target field, error information of the prediction signal for the image signal of the image of the encoding target field, information of the first motion vector, and information of a difference vector corresponding to the difference And a coding means for coding the moving picture. 9. The moving picture coding apparatus according to claim 1, wherein said minute value is set to 1. 10. The moving picture coding apparatus according to claim 5, wherein said coding means includes means for performing variable length coding on said error information. 11. The apparatus according to claim 6, wherein said encoding means includes means for performing variable length encoding on the error information, the information on the first motion vector, and the information on the difference vector. Video encoding device. 12. A method according to claim 1, wherein the first and second images between the first and second reference field images and the image of the encoding target field are successive.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of A mode for encoding error information of a prediction signal, wherein a vector having a direction closest to the first motion vector and a second motion vector among a plurality of motion vectors that can be the second motion vector A moving image encoding method, wherein the mode is used only when the magnitude of the difference is equal to or smaller than a minute value. 13. A method according to claim 1, wherein the first and second images between the successive first and second reference field images and the image of the encoding target field are arranged.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of A mode for encoding error information of a prediction signal, wherein the second
The magnitude of the difference between the vector closest to the first motion vector and the second motion vector among a plurality of motion vectors that can be the
A moving picture coding method having a mode used only in a range . 14. A method according to claim 1, further comprising the steps of:
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of Error information of a prediction signal, information of the first motion vector, and a vector whose direction is closest to the first motion vector among a plurality of motion vectors that can be the second motion vector, and the second motion vector And a mode for encoding information of a difference vector corresponding to a difference between the moving image coding method and the moving image coding method, wherein the mode is used only when the magnitude of the difference is equal to or smaller than a small value. 15. A method according to claim 1, wherein the first and second images between the successive images of the first and second reference fields and the image of the field to be encoded are successively arranged.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of Error information of a prediction signal, information of the first motion vector, and a vector whose direction is closest to the first motion vector among a plurality of motion vectors that can be the second motion vector, and the second motion vector Is a mode for encoding information of a difference vector corresponding to the difference between
A moving picture coding method comprising a mode used only in a range where the magnitude of the difference is not more than a minute value. 16. A method according to claim 1, wherein the first and second images between the successive images of the first and second reference fields and the image of the field to be encoded are successive.
Generating a prediction signal for the image of the encoding target field from a reference signal on the image of the first and second reference fields specified by the first and second motion vectors, respectively. Generating the error information of the prediction signal with respect to the image signal of the image of the field to be encoded; and generating the first motion vector among the plurality of motion vectors that can be the second motion vector. A coding step of coding the error information only when a magnitude of a difference between a vector having the closest direction and the second motion vector is a minute value or less. 17. The method according to claim 1, wherein the first and second images between the successive first and second reference field images and the image of the encoding target field are successively arranged.
Generating a prediction signal for the image of the encoding target field from a reference signal on the image of the first and second reference fields specified by the first and second motion vectors, respectively. Generating the error information of the prediction signal with respect to the image signal of the image of the field to be encoded; and generating the first motion vector among the plurality of motion vectors that can be the second motion vector. Generating a difference vector corresponding to the difference between the vector whose direction is closest to the second motion vector; and the error information and the information of the first motion vector only when the magnitude of the difference is a small value or less. And an encoding step of encoding the information of the difference vector, respectively. Determining a more reference signals on the image of the first reference field specified in the first motion vector between the 18. An image of the image and the encoding target field of a first reference field, A motion vector between an image of a second reference field contiguous to the first reference field and an image of the current field, wherein an image signal on the second reference field and a current field before <br/> Symbol second motion vector magnitude of the difference is that Do the following minute value to the nearest vector direction in the first motion vector exist between the images
Determining a reference signal on the image of the second reference field specified by the second motion vector only when the image is present; and determining the code from the reference signal on the image of the first and second reference fields. Generating a prediction signal for the image of the encoding target field; and encoding the error information of the prediction signal for the image signal of the image of the encoding target field. Method. Determining a more reference signals on the image of the first reference field specified in the first motion vector between the 19. An image of the image and the encoding target field of a first reference field, A motion vector between an image of a second reference field contiguous to the first reference field and an image of the current field, wherein an image signal on the second reference field and a current field before <br/> Symbol second motion vector magnitude of the difference is that Do the following minute value to the nearest vector direction in the first motion vector exist between the images
Determining a reference signal on the image of the second reference field specified by the second motion vector only when the image is present; and determining the code from the reference signal on the image of the first and second reference fields. Generating a prediction signal for the image of the encoding target field; error information of the prediction signal for the image signal of the image of the encoding target field; information of the first motion vector; and information of a difference vector corresponding to the difference. And a coding step of coding (a). 20. The moving picture coding method according to claim 12, wherein the minute value is set to 1. 21. The moving picture coding method according to claim 16, wherein said coding step includes a step of performing variable length coding on said error information. 22. The method according to claim 17, wherein the encoding step includes a step of performing variable length encoding of the error information, the information of the first motion vector, and the information of the difference vector. Video encoding method. 23. A method according to claim 1, wherein the first and second images between the first and second reference field images and the image of the encoding target field are successive.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of has a mode for encoding error information of the prediction signal, the first motion vector enlarged converted or reduced in terms of
The obtained image of the second reference field and the encoding target
Before the direction closest to the vector to the image of the field
The image on the second reference field and the field to be encoded
A moving image encoding apparatus which uses the mode only when the magnitude of the difference between the motion vector between the second motion vector and the second motion vector is smaller than a minute value. 24. A method according to claim 1, wherein the first and second images between the first and second reference field images and the image of the encoding target field are successive.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of A mode for encoding error information of a prediction signal, wherein the first
Obtained by converting the expanded terms or reduced motion vector
The image of the second reference field and the field to be encoded
The second closest in direction to the vector to the image of the second
Image on reference field and image on encoding target field
A moving image encoding apparatus having a mode used only in a range in which the magnitude of the difference between the motion vector between the second motion vector and the second motion vector is a minute value or less. 25. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of error information of the prediction signal, the first motion vector information, and the first motion vector enlarged converted or reduced in terms of
The obtained image of the second reference field and the encoding target
Before the direction closest to the vector to the image of the field
The image on the second reference field and the field to be encoded
And a mode for encoding information of a difference vector corresponding to a difference between a motion vector between the image and the second motion vector, and the mode is used only when the magnitude of the difference is a minute value or less. A moving picture coding apparatus characterized by using: 26. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of error information of the prediction signal, the first motion vector information, and the first motion vector enlarged converted or reduced in terms of
The obtained image of the second reference field and the encoding target
Before the direction closest to the vector to the image of the field
The image on the second reference field and the field to be encoded
A mode in which information of a difference vector corresponding to a difference between a motion vector between the image and the second motion vector is encoded, and is used only in a range where the magnitude of the difference is a minute value or less. A moving picture coding apparatus having a mode to be performed. 27. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Means for generating a motion vector of: a reference signal on an image of the first and second reference fields specified by the first and second motion vectors, respectively; Means for generating; means for generating error information of the prediction signal with respect to an image signal of an image of the field to be encoded; and magnification conversion or reduction conversion of the first motion vector.
The obtained image of the second reference field and the encoding target
Before the direction closest to the vector to the image of the field
The image on the second reference field and the field to be encoded
Encoding means for encoding the error information only when the magnitude of the difference between the motion vector between the motion vector and the second motion vector is a minute value or less. Device. 28. A method according to claim 28, wherein the first and second images between the successive images of the first and second reference fields and the image of the field to be encoded are successively arranged.
Means for generating a motion vector of: a reference signal on an image of the first and second reference fields specified by the first and second motion vectors, respectively; Means for generating; means for generating error information of the prediction signal with respect to an image signal of an image of the field to be encoded; and magnification conversion or reduction conversion of the first motion vector.
The obtained image of the second reference field and the encoding target
Before the direction closest to the vector to the image of the field
The image on the second reference field and the field to be encoded
Means for generating a difference vector corresponding to a difference between a motion vector between the first image and the second motion vector, the error information only when the magnitude of the difference is a minute value or less, A moving picture coding apparatus comprising: coding means for coding information on a motion vector and information on the difference vector. Means for determining more reference signals on the image of the first reference field specified in the first motion vector between the 29. The image of the first image and the encoding target field of the reference field, A motion vector between an image of a second reference field continuous with the first reference field and an image of the current field, wherein the first motion vector is enlarged.
The second reference feature obtained by converting or reduced terms
Vector between the field image and the image of the encoding target field.
Image on the second reference field closest to the
The only in the presence of a second motion vector magnitude is that Do the following minute value of the difference between the motion vector and the second motion vector between the images of the image and the encoding target field
Means for determining a reference signal on the image of the second reference field specified by a second motion vector; and determining a reference signal on the image of the first and second reference fields from the reference signal on the image of the encoding target field. A moving picture coding apparatus comprising: means for generating a prediction signal; and coding means for coding error information of the prediction signal with respect to an image signal of an image of the field to be coded. And 30. A means for obtaining the first reference signal on the image of the reference field further specified in the first motion vector between the image of the image and the encoding target field of a first reference field, A motion vector between an image of a second reference field continuous with the first reference field and an image of the current field, wherein the first motion vector is enlarged.
The second reference feature obtained by converting or reduced terms
Vector between the field image and the image of the encoding target field.
Image on the direction nearest the second reference field on Torr
The only in the presence of a second motion vector magnitude is that Do the following minute value of the difference between the motion vector and the second motion vector between the images of the image and the encoding target field
Means for determining a reference signal on the image of the second reference field specified by a second motion vector; and determining a reference signal on the image of the first and second reference fields from the reference signal on the image of the encoding target field. Means for generating a prediction signal; encoding information for encoding error information of the prediction signal with respect to an image signal of an image of the field to be encoded; information for the first motion vector; and information for a difference vector corresponding to the difference. A moving image encoding apparatus comprising: 31. The moving picture coding apparatus according to claim 23, wherein said minute value is set to 1. 32. A moving picture coding apparatus according to claim 27, wherein said coding means includes means for performing variable length coding on said error information. 33. The encoding apparatus according to claim 28, wherein said encoding means includes means for performing variable-length encoding on the error information, the information on the first motion vector, and the information on the difference vector. Video encoding device. 34. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of has a mode for encoding error information of the prediction signal, the first motion vector enlarged converted or reduced in terms of
The obtained image of the second reference field and the encoding target
Forward close most to the vector between the field image
The image on the second reference field and the field to be encoded
A moving image encoding method, wherein the mode is used only when the magnitude of a difference between a motion vector between the image and the second motion vector is a small value or less. 35. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of A mode for encoding error information of a prediction signal, wherein the first
Obtained by converting the expanded terms or reduced motion vector
The image of the second reference field and the field to be encoded
The second closest in direction to the vector to the image of the second
Image on reference field and image on encoding target field
A mode in which the magnitude of the difference between the motion vector between the second motion vector and the second motion vector is smaller than or equal to a small value. 36. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of error information of the prediction signal, the first motion vector information, and the first motion vector enlarged converted or reduced in terms of
The obtained image of the second reference field and the encoding target
Before the direction closest to the vector to the image of the field
The image on the second reference field and the field to be encoded
And a mode for encoding information of a difference vector corresponding to a difference between a motion vector between the image and the second motion vector, and the mode is used only when the magnitude of the difference is a minute value or less. A moving picture coding method characterized by using: 37. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of error information of the prediction signal, the first motion vector information, and the first motion vector enlarged converted or reduced in terms of
The obtained image of the second reference field and the encoding target
Before the direction closest to the vector to the image of the field
The image on the second reference field and the field to be encoded
A mode in which information of a difference vector corresponding to a difference between a motion vector between the image and the second motion vector is encoded, and is used only in a range where the magnitude of the difference is a minute value or less. A moving image encoding method having a mode to be performed. 38. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from a reference signal on the image of the first and second reference fields specified by the first and second motion vectors, respectively. and generating, said the step of generating the error information of the prediction signal, the first motion vector enlarged converted or reduced converted to the image signal of the encoding target fields of image
The obtained image of the second reference field and the encoding target
Before the direction closest to the vector to the image of the field
The image on the second reference field and the field to be encoded
A coding step of coding the error information only when the magnitude of the difference between the motion vector between the motion vector and the second motion vector is a small value or less. Method. 39. First and second images between a continuous first and second reference field image and an encoding target field image.
Generating a prediction signal for the image of the encoding target field from a reference signal on the image of the first and second reference fields specified by the first and second motion vectors, respectively. and generating, said the step of generating the error information of the prediction signal, the first motion vector enlarged converted or reduced converted to the image signal of the encoding target fields of image
The obtained image of the second reference field and the encoding target
Before the direction closest to the vector to the image of the field
The image on the second reference field and the field to be encoded
Generating a difference vector corresponding to a difference between a motion vector between the first image and the second motion vector, and the error information, the first information only when the magnitude of the difference is a minute value or less. A coding step of coding the information of the motion vector and the information of the difference vector, respectively. Determining a more reference signals on the image of the first reference field specified in the first motion vector between 40. images of the image and the encoding target field of a first reference field, A motion vector between an image of a second reference field continuous with the first reference field and an image of the current field, wherein the first motion vector is enlarged.
The second reference feature obtained by converting or reduced terms
Vector between the field image and the image of the encoding target field.
Image on the second reference field closest to the
The only in the presence of a second motion vector magnitude is that Do the following minute value of the difference between the motion vector and the second motion vector between the images of the image and the encoding target field
Obtaining a reference signal on the image of the second reference field specified by a second motion vector; and determining a reference signal on the image of the first and second reference fields from the reference signal on the image of the encoding target field. A moving image encoding method, comprising: generating a prediction signal; and encoding an error information of the prediction signal with respect to an image signal of an image of the encoding target field. Determining a more reference signals on the image of the first reference field specified in the first motion vector between 41. images of the image and the encoding target field of a first reference field, A motion vector between an image of a second reference field continuous with the first reference field and an image of the current field, wherein the first motion vector is enlarged.
The second reference feature obtained by converting or reduced terms
Vector between the field image and the image of the encoding target field.
Image on the second reference field closest to the
The only in the presence of a second motion vector magnitude is that Do the following minute value of the difference between the motion vector and the second motion vector between the images of the image and the encoding target field
Obtaining a reference signal on the image of the second reference field specified by a second motion vector; and determining a reference signal on the image of the first and second reference fields from the reference signal on the image of the encoding target field. Generating a prediction signal; and encoding encoding error information of the prediction signal with respect to the image signal of the image of the encoding target field, information of the first motion vector, and information of a difference vector corresponding to the difference. And a moving image encoding method. 42. The moving picture coding method according to claim 34, wherein said minute value is set to one. 43. The moving picture coding method according to claim 38, wherein said coding step includes a step of performing variable length coding on said error information. 44. The method according to claim 39, wherein the encoding step includes a step of performing variable length encoding on the error information, the information on the first motion vector, and the information on the difference vector. Video encoding method. 45. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of has a mode for encoding error information of the prediction signal, said first motion vector of the second reference field
Replace between the image and the image of the encoding target field
The second reference file closest in direction to the vector obtained.
Between the field image and the image of the encoding target field.
Video encoding apparatus magnitude of the difference between the and the can vector the second motion vectors, characterized by using only the mode when the following minute value. 46. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of A mode for encoding error information of a prediction signal, wherein the first
Images and marks the motion vector and the second reference field
Obtained by replacing with the image of the encoding target field
Vector of the second reference field closest in direction to the vector
A mode in which a difference between a motion vector between an image and an image of an encoding target field and the second motion vector is used only in a range of a small value or less. Video encoding device. 47. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of The error information of the prediction signal, the information of the first motion vector, and the first motion vector are used for the second reference field .
Replace between the image and the image of the encoding target field
The second reference file closest in direction to the vector obtained.
Between the field image and the image of the encoding target field.
Has a mode of each encoding information of the difference vector corresponding to the difference between the and the can vector the second motion vector, and wherein the magnitude of the difference is used only the mode when the following minute value Video encoding device. 48. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of The error information of the prediction signal, the information of the first motion vector, and the first motion vector are used for the second reference field .
Replace between the image and the image of the encoding target field
The second reference file closest in direction to the vector obtained.
Between the field image and the image of the encoding target field.
A mode for each encoding information of the difference vector corresponding to the difference between the and the can vector the second motion vector, to have a mode size of the difference is used only within the following minute value A moving picture coding apparatus characterized by the following. 49. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Means for generating a motion vector of: a reference signal on an image of the first and second reference fields specified by the first and second motion vectors, respectively; Means for generating, error information of the prediction signal with respect to the image signal of the image of the field to be encoded, and means for generating the first motion vector of the second reference field .
Replace between the image and the image of the encoding target field
The second reference file closest in direction to the vector obtained.
Between the field image and the image of the encoding target field.
Video encoding apparatus magnitude of the difference between the and the can vector the second motion vectors and having a coding means for coding the error information only when the following minute value. 50. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Means for generating a motion vector of: a reference signal on an image of the first and second reference fields specified by the first and second motion vectors, respectively; Means for generating, error information of the prediction signal with respect to the image signal of the image of the field to be encoded, and means for generating the first motion vector of the second reference field .
Replace between the image and the image of the encoding target field
The second reference file closest in direction to the vector obtained.
Between the field image and the image of the encoding target field.
Means and the error information only when the magnitude is less than a minute value of the difference information and the difference between the first motion vector to produce a difference vector corresponding to the difference between the and the can vector the second motion vector A moving picture coding apparatus comprising: coding means for coding vector information. Means for determining more reference signals on the image of the first reference field specified in the first motion vector between the 51. image of the first image and the encoding target field of the reference field, A motion vector between an image of a second reference field contiguous to the first reference field and an image of the field to be encoded, wherein the first motion vector is an image of the second reference field ; Of the field to be encoded
The direction of the vector obtained by replacing
The image of the second reference field and the encoding target
If there is a second motion vector magnitude is that Do the following minute value of the difference between the motion vector between the image of Rudo
Only the second specified by the second motion vector
Means for obtaining a reference signal on the image of the reference field of the above, means for generating a prediction signal for the image of the field to be encoded from the reference signals on the images of the first and second reference fields, Encoding means for encoding error information of the prediction signal with respect to an image signal of a field image. And 52. means for determining a first reference signal on the image of the reference field further specified in the first motion vector between the image of the image and the encoding target field of a first reference field, A motion vector between an image of a second reference field contiguous to the first reference field and an image of the field to be encoded, wherein the first motion vector is an image of the second reference field ; Of the field to be encoded
The direction of the vector obtained by replacing
The image of the second reference field and the encoding target
If there is a second motion vector magnitude is that Do the following minute value of the difference between the motion vector between the image of Rudo
Only the second specified by the second motion vector
Means for obtaining a reference signal on the image of the reference field, means for generating a prediction signal for the image of the field to be encoded from the reference signals on the image of the first and second reference fields, Encoding means for encoding error information of the prediction signal with respect to an image signal of a field image, information of the first motion vector, and information of a difference vector corresponding to the difference. Device. 53. The moving picture coding apparatus according to claim 45, wherein said minute value is set to 1. 54. The moving picture coding apparatus according to claim 49, wherein said coding means includes means for performing variable length coding on said error information. 55. The encoding apparatus according to claim 50, wherein the encoding means includes means for performing variable length encoding on the error information, the information on the first motion vector, and the information on the difference vector. Video encoding device. 56. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of has a mode for encoding error information of the prediction signal, said first motion vector of the second reference field
Replace between the image and the image of the encoding target field
The second reference file closest in direction to the vector obtained.
Between the field image and the image of the encoding target field.
Moving picture coding method magnitude of the difference between the and the can vector the second motion vectors, characterized by using only the mode when the following minute value. 57. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of A mode for encoding error information of a prediction signal, wherein the first
Images and marks the motion vector and the second reference field
Obtained by replacing with the image of the encoding target field
Vector of the second reference field closest in direction to the vector
A mode in which a difference between a motion vector between an image and an image of an encoding target field and the second motion vector is used only in a range of a small value or less. Moving image encoding method. 58. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of The error information of the prediction signal, the information of the first motion vector, and the first motion vector are used for the second reference field .
Instead placed between the images of the image and the encoding target field
The second reference file closest in direction to the vector obtained.
Between the field image and the image of the encoding target field.
Has a mode of each encoding information of the difference vector corresponding to the difference between the and the can vector the second motion vector, and wherein the magnitude of the difference is used only the mode when the following minute value Moving image encoding method. 59. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of The error information of the prediction signal, the information of the first motion vector, and the first motion vector are used for the second reference field .
Replace between the image and the image of the encoding target field
The second reference file closest in direction to the vector obtained.
Between the field image and the image of the encoding target field.
A mode for each encoding information of the difference vector corresponding to the difference between the and the can vector the second motion vector, to have a mode size of the difference is used only within the following minute value A moving picture coding method characterized by the following. 60. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from a reference signal on the image of the first and second reference fields specified by the first and second motion vectors, respectively. Generating the error information of the prediction signal with respect to the image signal of the image of the field to be encoded; and generating the first motion vector of the second reference field .
Replace between the image and the image of the encoding target field
The second reference file closest in direction to the vector obtained.
Between the field image and the image of the encoding target field.
Moving picture coding method magnitude of the difference between the and the can vector the second motion vector and having an encoding step of encoding the error information only when the following minute value. 61. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from a reference signal on the image of the first and second reference fields specified by the first and second motion vectors, respectively. Generating the error information of the prediction signal with respect to the image signal of the image of the field to be encoded; and generating the first motion vector of the second reference field .
Replace between the image and the image of the encoding target field
The second reference file closest in direction to the vector obtained.
Between the field image and the image of the encoding target field.
Step a, the error information only when the magnitude is less than a minute value of the difference information and the difference between the first motion vector to produce a difference vector corresponding to the difference between the and the can vector the second motion vector A coding step of coding vector information, respectively. Determining a more reference signals on the image of the first reference field specified in the first motion vector between the 62. An image of the image and the encoding target field of a first reference field, A motion vector between an image of a second reference field contiguous to the first reference field and an image of the field to be encoded, wherein the first motion vector is an image of the second reference field ; Of the field to be encoded
The direction of the vector obtained by replacing
The image of the second reference field and the encoding target
If there is a second motion vector magnitude is that Do the following minute value of the difference between the motion vector between the image of Rudo
The only specified more to the second motion vector second
Obtaining a reference signal on the image of the reference field of the following; generating a prediction signal for the image of the field to be encoded from the reference signals on the images of the first and second reference fields; A coding step of coding error information of the prediction signal with respect to an image signal of a field image. Determining a more reference signals on the image of the first reference field specified in the first motion vector between 63. images of the image and the encoding target field of a first reference field, A motion vector between an image of a second reference field contiguous to the first reference field and an image of the field to be encoded, wherein the first motion vector is an image of the second reference field ; Of the field to be encoded
The direction of the vector obtained by replacing
The image of the second reference field and the encoding target
If there is a second motion vector magnitude is that Do the following minute value of the difference between the motion vector between the image of Rudo
Only the second specified by the second motion vector
Obtaining a reference signal on the image of the reference field of the reference signal; generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields; An encoding step of encoding error information of the prediction signal with respect to an image signal of an image of an encoding target field, information of the first motion vector, and information of a difference vector corresponding to the difference. Video encoding method. 64. The moving picture coding method according to claim 56, wherein said minute value is set to 1. 65. The moving picture coding method according to claim 61, wherein said coding step includes a step of performing variable length coding on said error information. 66. The coding method according to claim 63, wherein the coding step includes a step of performing variable length coding on the error information, the information on the first motion vector, and the information on the difference vector. Video encoding method. 67. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of A mode for encoding error information of a prediction signal, wherein a ratio of a time interval between the encoding target field and the first reference field and a time interval between the encoding target field and the second reference field is determined. The second reference obtained by converting the magnitude of the first motion vector accordingly.
Between the reference field image and the encoding target field image
The second reference field closest in direction to the vector between
A motion vector between the motion vector and the second motion vector between the motion vector image and the image of the encoding target field, the mode being used only when the magnitude of the difference between the motion vector and the second motion vector is a minute value or less. Device. 68. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the image signal on the image of the first and second reference fields respectively specified by the motion vector of A mode for encoding error information of a prediction signal, wherein a ratio of a time interval between the encoding target field and the first reference field to a time interval between the encoding target field and the second reference field is determined. The second reference field obtained by converting the magnitude of the first motion vector accordingly.
Vector between the field image and the image of the encoding target field.
Image of the second reference field closest in direction to the tor
A moving image encoding apparatus having a mode in which the difference between the motion vector between the image and the image of the encoding target field and the second motion vector is used only in a range where the magnitude of the difference is a minute value or less. . 69. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of Error information of a prediction signal, information of the first motion vector, and a time interval between the encoding target field and the first reference field and a time interval between the encoding target field and the second reference field. The second reference obtained by converting the magnitude of the first motion vector according to a ratio
Between the image of the field and the image of the field to be encoded
The second reference field closest in direction to the vector
And a mode for encoding information of a difference vector corresponding to a difference between a motion vector between the image of the image and the image of the encoding target field and the second motion vector. A moving image coding apparatus characterized in that the mode is used only when the size of the image is not more than a minute value. 70. A method according to claim 1, wherein the first and second reference fields are consecutively arranged between the first and second reference field images and the encoding target field image.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of Error information of a prediction signal, information of the first motion vector, and a time interval between the encoding target field and the first reference field and a time interval between the encoding target field and the second reference field. The second reference obtained by converting the magnitude of the first motion vector according to the ratio.
Between the reference field image and the encoding target field image
The second reference field closest in direction to the vector between
A mode in which information of a difference vector corresponding to a difference between a motion vector between the image of the target and the image of the field to be encoded and the second motion vector is encoded. A moving image encoding apparatus having a mode used only in a range where the size of the image is less than or equal to a minute value. 71. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Means for generating a motion vector of: a reference signal on an image of the first and second reference fields specified by the first and second motion vectors, respectively; Means for generating; means for generating error information of the prediction signal with respect to an image signal of an image of the field to be coded; time interval between the field to be coded and the first reference field; and the field to be coded. The second reference field obtained by converting the magnitude of the first motion vector in accordance with the ratio of the time interval between the first motion vector and the second reference field.
Between the reference field image and the encoding target field image
The second reference field closest in direction to the vector between
Encoding means for encoding the error information only when the magnitude of the difference between the motion vector between the image of the target and the image of the field to be encoded and the second motion vector is not more than a minute value. And a moving image encoding apparatus. 72. First and second images between consecutive images of the first and second reference fields and images of the current field to be encoded.
Means for generating a motion vector of: a reference signal on an image of the first and second reference fields specified by the first and second motion vectors, respectively; Means for generating; means for generating error information of the prediction signal with respect to an image signal of an image of the field to be coded; time interval between the field to be coded and the first reference field; and the field to be coded. The second reference field obtained by converting the magnitude of the first motion vector in accordance with the ratio of the time interval between the first motion vector and the second reference field.
Between the reference field image and the encoding target field image
The second reference field closest in direction to the vector between
Means for generating a difference vector corresponding to a difference between a motion vector between the image of the target and the image of the field to be encoded and the second motion vector; A moving picture coding apparatus comprising: coding means for coding the error information, the first motion vector information, and the difference vector information only in the following cases. 73. The moving picture coding apparatus according to claim 67, wherein said minute value is set to 1. 74. The moving picture coding apparatus according to claim 71, wherein said coding means includes means for performing variable length coding on said error information. 75. A moving image according to claim 72, wherein said encoding means includes means for performing variable length encoding on said error information, said first motion vector information and said difference vector information. Image coding device. 76. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from a reference signal on the image of the first and second reference fields specified by the motion vector of A mode for encoding the error information of the signal, according to a ratio of a time interval between the encoding target field and the first reference field and a time interval between the encoding target field and the second reference field. The second reference obtained by converting the magnitude of the first motion vector
Between the reference field image and the encoding target field image
The second reference field closest in direction to the vector between
The mode is used only when the magnitude of the difference between the motion vector between the image of the target and the image of the field to be encoded and the second motion vector is a small value or less. Moving image encoding method. 77. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of A mode for encoding error information of a prediction signal, wherein a ratio of a time interval between the encoding target field and the first reference field to a time interval between the encoding target field and the second reference field is determined. The second reference field obtained by converting the magnitude of the first motion vector accordingly.
Vector between the field image and the image of the encoding target field.
Image of the second reference field closest in direction to the tor
A moving image encoding method having a mode used only in a range in which the magnitude of the difference between the motion vector between the image and the image of the encoding target field and the second motion vector is a minute value or less. . 78. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of An error signal of a prediction signal, information of the first motion vector, and a time interval between the encoding target field and the first reference field and a time interval between the encoding target field and the second reference field. The second reference obtained by converting the magnitude of the first motion vector according to the ratio.
Between the reference field image and the encoding target field image
The second reference field closest in direction to the vector between
And an encoding mode for encoding information of a difference vector corresponding to a difference between a motion vector between the image of the target and the image of the encoding target field and the second motion vector. And using the mode only when the magnitude of the difference is equal to or smaller than a minute value. 79. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from the reference signal on the image of the first and second reference fields respectively specified by the motion vector of An error signal of a prediction signal, information of the first motion vector, and a time interval between the encoding target field and the first reference field and a time interval between the encoding target field and the second reference field. The second reference obtained by converting the magnitude of the first motion vector according to the ratio.
Between the reference field image and the encoding target field image
The second reference field closest in direction to the vector between
A mode in which information of a difference vector corresponding to a difference between a motion vector between the image of the target and the image of the field to be encoded and the second motion vector is encoded. A moving image encoding method having a mode used only in a range where the size of the image is not more than a minute value. 80. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from a reference signal on the image of the first and second reference fields specified by the first and second motion vectors, respectively. Generating, the step of generating error information of the prediction signal with respect to the image signal of the image of the field to be coded, the time interval between the field to be coded and the first reference field, and the field to be coded The second reference field obtained by converting the magnitude of the first motion vector in accordance with the ratio of the time interval between the first motion vector and the second reference field.
Between the reference field image and the encoding target field image
The second reference field closest in direction to the vector between
Encoding the error information only when the magnitude of the difference between the motion vector between the image of the target and the image of the encoding target field and the second motion vector is a small value or less. A moving picture coding method comprising: 81. First and second images between successive images of the first and second reference fields and images of the current field to be encoded.
Generating a prediction signal for the image of the encoding target field from a reference signal on the image of the first and second reference fields specified by the first and second motion vectors, respectively. Generating, the step of generating error information of the prediction signal with respect to the image signal of the image of the field to be coded, the time interval between the field to be coded and the first reference field, and the field to be coded The second reference field obtained by converting the magnitude of the first motion vector in accordance with the ratio of the time interval between the first motion vector and the second reference field.
Between the reference field image and the encoding target field image
The second reference field closest in direction to the vector between
Generating a difference vector corresponding to a difference between a motion vector between the image of the target and the image of the field to be encoded and the second motion vector; Encoding the error information, the information on the first motion vector, and the information on the difference vector only in the following cases: 82. The moving picture coding method according to claim 76, wherein said minute value is set to one. 83. A moving picture coding method according to claim 80, wherein said coding step includes a step of performing variable length coding on said error information. 84. The moving picture according to claim 81, wherein said encoding step includes a step of performing variable length encoding on said error information, said first motion vector information and said difference vector information. Image coding method.