JPS63221782A - Movement correction sub-sample interpolation system - Google Patents

Abstract

PURPOSE:To suppress the deterioration of resolution at movement correction by predicting the movement information between fields parted by one field to a half of the moving vector of the sender side, applying movement correction and using the inter-field interpolation filter. CONSTITUTION:A sample rate signal of 16.2MHz inputted from a video input terminal 1 and an output signal of a 2nd movement correction filed memory 5 are switched in the timing of 32.4MHz sub-sample by a switch 2, inputted to a 1st movement correction field memory 4 and inputted to the memory 5. The signal by one frame is stored in the memories 4, 5 and the movement correction is applied by using the inter-field predicting vector halved into the movement vector at the sender side by a divider 3. Then the missing sample point is interpolated by using the inter-field interpolation filter 6 based on the signal passing through the switch 2, and the output signal of the memory 4.

Description

[Detailed Description of the Invention] [Field of Industrial Application] This invention is applied to image processing in a high-definition television receiver using the MIJSE system, and calculates the amount of motion correction between each field from the transmitted motion information. The present invention relates to a motion compensation subsample interpolation method that calculates and interpolates missing points between fields.

[Prior art] This type of motion compensation subsample interpolation method that has been proposed in the past is based on the ``new technology for high-definition television'' announced by the Broadcasting Technology Research Institute at the Sowa Memorial Lecture held in June 1980. Transmission method (M[JSE) There is a method described in the preliminary document called J. Fig. 8 shows the configuration of the motion compensation subsample interpolation method.

In FIG. 8, (1) is a video input terminal that inputs a video signal transmitted at a sample rate of 16.2 MHz;
(2) is a switch that switches between the video signal input to the L input terminal (1) and the output signal of a motion correction field memory (5), which will be described later, at a subsample timing of 32.4 Mllz. (10) is a non-motion compensated field memory, which passes through the above switch (2): 12.414H
A signal for one frame at a sample rate of z is stored. (
5) is a motion compensation field memory, which is the 32.4 MH output from the non-motion compensation field memory (10).
A signal for one frame at a sample rate of z is stored, and motion correction is performed using a motion vector.

(6) is an interfield interpolation filter, which combines the output signal of the non-motion compensation field memory (10) with the switch (2).
Interpolate missing sample points based on the signal that has passed through. (
7) is the in-field filter, and the above switch (2)
), the missing sample points are interpolated based on only the signals that have passed through. (8) is a switch that is connected to the upper contact (8X) when motion compensation is not performed and passes the output signal of the interfield interpolation filter (6), and when motion compensation is performed or motion detection is performed, the switch is connected to the upper contact (8X). It is connected to the side contact (8y) and passes the output signal of the in-field sleeve filter (7). (9) is a missing sample point or interpolated 6
This is a video output terminal that outputs a signal that has a sun frequency of 4.8M1lz and passes through the two-legged switch (8).

Next, the operation of the first configuration will be explained.

The transmission method for high-definition television is Sagniquist sampling, which goes around four fields. The bandwidth is 8.1 bars.

When the first field signal is input to the video input terminal (1), the non-motion compensation field memory (10) stores the second and fourth field signals from the previous round, while the motion compensation field memory (5) stores the signals of the first field and the third field from the previous round. When a motion vector exists due to camera panning,
The contents of the motion correction field memory (5) move two-dimensionally according to the vector. At this time, if the transmitted motion vector is the amount of motion correction between fields separated by one frame, the motion correction content of the motion correction field memory (5) is the first field input to the video input terminal (1). This is done based on the signal.

The switch (2) switches at the timing of 32.4 MHz sub-samples, and the phase is reversed for each field.
It is also reversed depending on the motion vector. Therefore, in the above case, the signal of the first field input from the video input terminal (1) and the signal of the third field one cycle before motion correction are passed through the switch (2).

The signal passed through the switch (2) and the output signal of the non-motion compensated field memory (10) are input to an interfield interpolation filter (6) to perform interfield interpolation.

When motion compensation is not performed, that is, when the switch (8) is connected to the upper contact (8x), the video output terminal (
9) outputs a video signal with a sample rate of 64.8 MHz subjected to interfield interpolation.

In addition, the field signal that has passed through the switch (2) is
It is input to the intra-field interpolation filter (7), performs the intra-field interpolation, and is connected to the switch (8) or the lower contact (8y).
), the video output terminal (9) outputs a video signal with a sample rate of 64.8 MHz, which is inter-field spaced.

When the a field is input to the video input terminal (1) without motion compensation 1F being performed, the states of the signals 9a to 9e in Figure JS8 are as shown in Figure 9.

However, if the signal is expressed as a field, it is a+.

Ill, C+, (L +aZ +bL+CZ +d
! −” ”” A in FIG. 9 is an interpolation function f8, which is interfield processing, and b+ +CI
, d, +aL.

When motion correction is performed and the az field is input to the video input terminal (1), the states of the signals lOa to Inc in FIG. 8 are as shown in FIG. 1O, and in FIG. 1O,
A horizontal line above the symbol indicates that motion compensation has been performed, and B
is an interpolation function that is an intra-field process.
Indicates the interpolated value interpolated from l.

[Problems to be Solved by the Invention] When using the conventional motion compensation subsample interpolation method as described above, the relationship between adjacent fields is not taken into consideration at all when performing motion compensation, resulting in missing samples. Interpolation of points must be done within the field by means of an intrafield spacing filter. As a result, there was a problem in that the resolution during motion correction was reduced.

This invention was made to solve the above-mentioned problems, and it is possible to calculate the motion vector between each field from the transmitted motion vector and perform motion compensation in the same way as when motion compensation is not performed. Another object of the present invention is to provide a motion compensation subsample interpolation method that can suppress a decrease in resolution by using an interfield interpolation filter.

Means for Solving Problem U The motion compensation subsample interpolation method according to the present invention is based on motion information between transmitted fields spaced apart by one frame, on the receiving side. The motion information of If is approximated by 1/2 on the sending side, and the motion of all signals V) from the first field to the fourth field is calculated based on the measured motion information. The present invention is characterized in that it is configured to perform inter-field interpolation by correction.

[Operations] According to this invention, by taking advantage of the fact that there is no sudden change in the transmitted motion vector, ■ approximating the motion information between past fields separated by fields to I/2 of the motion vector on the transmission side. By making predictions, it is possible to always perform interpolation using an interfield interpolation filter, thereby suppressing a decrease in resolution during motion correction.

[Example 1] From now on, an example of this IJJ will be described based on the drawings.

FIG. 1 is a block diagram showing a motion compensation subsample interpolation method according to an embodiment of the present invention, in which (1) represents 12 images transmitted at a sample rate of +6.2 The video input terminal (2) that inputs the signal is switched to output the video signal input to the video input terminal (1) and the output signal of the second motion correction field memory (5), which will be described later, at 32.4 MI'. (3) is a divider that switches at the timing of the IZ subsample, and divides the transmitted motion vector by 1/
Divide by 2. (4) is the first motion compensation field memory, and the 32,4 MIIZ passing through the switch (2) is
The signal for one frame at the sample rate of is stored, and motion correction is performed using the inter-field T-measured motion vector output from the divider (3).

(5) is a second motion compensation field memory;
32 output from the motion correction field memory (4) of
．． 4. The signal for one frame at the MIIZ sample rate is stored, and motion correction is performed using the predicted motion vector between fields, which is the output of the divider by 1 (3).

(6) is an interfield interpolation filter, and the above switch (
2) and the output signal of the motion correction field memory (4) of L foot temperature 1, the missing sample points are interpolated. (7) is an intra-field interpolation filter that interpolates missing sample points based only on the signal passing through the switch (2). (8) is the switch, usually the right side contact (8x)
is connected to pass the output signal of the interfield interpolation filter (6).

When motion is detected, it is connected to the lower contact point (8y) at the middle of the image and passes the output value of the above-mentioned intra-field sleeve filter (7). (9) is a central image output terminal which outputs a signal which interpolates the missing sample points to obtain a sample rate of 64.8 MIIZ and passes through the switch (8) shown in -.

Next, the operation of the above configuration will be explained.

In order to specifically explain the operation shown in FIG. 1, the second FA shows a motion vector when banging occurs, and in the figure, X represents the horizontal axis of the screen and y represents the vertical axis of the screen.

Each field of the video signal is represented by aI to d,
Expressed as a field, +** +*・;3n, ba
, co , do , at +C+, dl, az, bz,
It flows in the order of cZ, dL... The motion vectors between fields and fields separated by one frame are...
...At, B1. (,+,D+,8z,Bz,
It is expressed as...

Also, the inter-field motion vectors separated by one field, which are overwritten on the drawing, are...e...b Io, C1o,
It is expressed as dIo, ato...

The following relational expression holds between the inter-field motion vectors separated by one frame and the inter-field motion vectors separated by one frame.

atQ+b+o=A+b10+○IOII B+C1o+d+o=C+Here, consider only the relationship between the three fields 81 to c1, for example. However, the inter-field motion vectors separated by l frames are transmitted one per field as a digital signal of 5 bits in the horizontal direction and 3 bits in the vertical direction. In other words, if we describe the magnitude of the motion vector between fields separated by one frame in terms of pixel length, the horizontal component is in the range of -15 to +16, the vertical component is in the range of -3 to +4, and the horizontal component is in the range of -3 to +4. Both the component and the vertical component are the number g
It has dispersive components. To illustrate this, as shown in FIG. 3, the inter-field motion vectors separated by l frames are within the range surrounded by broken lines. However, the origin in FIG. 3 is an arbitrary pixel in the a1 field.

If the parallel movement of the image due to panning is smooth, a
The inter-field motion vectors separated by one field between the 1 field and the b1 field are within the range surrounded by the chain line in FIG. Here, the relative relationships among the three fields al to Cξ are changed in three ways and are shown in FIGS. 4 to 6.

First, FIG. 4 shows the case where the magnitude of the inter-field motion vector A1 transmitted one frame apart is at its maximum. At this time, the inter-field motion vector al that is separated by one field must also be maximum. Further, the inter-field motion vector 1) to separated by one field must also be within the range of the two-dot chain line or must have the maximum value.

Therefore, vector b·1o matches A1/2.

Next, FIG. 5 shows a case where the components of the inter-field motion vector A1 separated by one frame are (x, y)·(10, 2). As in FIG. 4, the inter-field motion vector ate that is separated by one field exists within the range surrounded by the chain line, and if the b1 field is at the point shown in the figure,
The inter-field motion vector b)o that is separated by one field exists within the range surrounded by the two-dot chain line. In fact, such movement by panning is possible because the C1 field is within the range enclosed by the dash-dotted line as shown.

From the following, when the relative positions of the a1 field and the 01 field are determined by the inter-field motion vector A+ transmitted l frames apart, the position of the b1 field with respect to those fields is within the shaded range. be. Moreover, if the parallel movement of the image due to panning is smooth, the position of the b1 field will be 3x centered on the midpoint of the positions of the a1 field and the C1 field, which are indicated by the black circles in Figure 5, even within the diagonally shaded range. There is a high possibility that it exists on the square lattice of 3. If you want, vector bIo
It is reasonable to substitute AI/2, and the error is 1
It is about the size of a pixel.

Next, FIG. 6 shows a case where the components of the inter-field motion vector A separated by one frame are (X, y) "(3, 1). Similarly to FIG. 4, the inter-field motion vector A separated by one field is If the vector ago lies within the range enclosed by the dashed line and is at the b+ field or the point shown, then 1
The inter-field motion vector o between the fields exists within the range surrounded by the two-dot chain line. Since the C1 field is within the dashed-dotted line as shown in the diagram, such movement due to panning is +i (function).The direction change of the motion vector between adjacent fields separated by one field exceeds the right angle. Since there is no such panning, when the relative positions of the aI field and the 01 field are determined by the inter-field motion vector A1 transmitted l frames apart, the position of the b1 field with respect to those fields is within the shaded range. Furthermore, if the translation of the image due to panning is smooth, the position of the b1 field will be centered at the midpoint of the positions of the a1 field and the C1 field, which are indicated by the black circles in Figure 6, even within the diagonally shaded range. There is a high possibility that it exists in a 2×2 square lattice.

Therefore, the vector bjo can be substituted with A1/2, but since the component of A1/2 in this case is not an integer value, the vector connecting one of the black circles of i6[A and the position of the C1 field is Used as a substitute for bro.

In this case as well, the error is about one pixel.

As described above, the inter-field motion vector bIo separated by 1 field can be predicted by approximating the integer part of Al/2 using the inter-field motion vector A separated by 1 frame, and the error is about 1 pixel or Similarly, it can be predicted as follows: ato[Do/2]bIo''F[Al/2]c+ogon[B+/2]. However, [] represents an integer part.

Moreover, between fields separated by this one field.

Errors in motion vector prediction do not affect subsequent predictions, so even if the error becomes large due to a temporary large vector change, predictions with smaller errors can be made again if the vector changes become more gradual.

Motion correction performed by prediction as described above will be explained with reference to FIG.

First, when a d1 field signal is input to the video input terminal f-(1) in FIG. 1, the first motion compensation field memory (
4), the signals of the 81 field and C1 field are stored, and the second motion correction field memory (5) stores the signals of the (lo field, b, field).At this time, when the motion vector B1 is input, ,;1.II calculator (
3), the motion vector 8I is divided by B+/2, and the acid 11 portion of the output is the interfield motion vector C+ separated by one field. This is the predicted value. The contents of the first motion correction field memory (4) are the f-measured vector [B
+/2] to move two-dimensionally to the video input terminal (1
), motion correction is performed based on the d1 field input manually.

On the other hand, the signals of the do field and b1 field stored in the second field memory (5) are
Before the field, there is a first motion compensation field memory (
4) Predicted vector [A
1/2], the first motion correction field memory (4) is based on the d) field.
Similarly, motion correction may be performed using the predicted vector [B+/2].

In this way, the C1 field is subjected to motion compensation based on the d1 field, b! field, C1 field,
The signal for four fields of the d+ field itself enters the interfield interpolation filter (6) to enable interfield interpolation. Normally, the switch (8) is connected to the upper contact (8x), and the inter-field interpolated signal is passed through, or in the case of video signals, motion is detected because missing sample points cannot be interpolated by inter-field interpolation. At this time, the switch (8) is connected to the lower contact (8y) in units of pixels, and the intra-field interpolation filter (7) performs intra-field interpolation only from the signal that has passed through the switch (2).

The sample rate of the signal whose missing sample points are interpolated by the interfield interpolation filter (6) or the intrafield interpolation filter (7) is 64.8 Mllz, and is output from the video output terminal (9).

In this motion compensation subsample interpolation method that calculates motion vectors between fields separated by one field using 1'-measurement, it is not only stable at the start and end of panning, but also has a Even if there is a temporal error, it will not leave any negative effects.

It is also stable even in special cases where the scene changes during panning or the image changes during panning. Furthermore, since the motion vector changes gradually due to panning, this method is very good.

711 shows the states of the signals 7a to 7g in Figure 0'51 when the d1 field signal is input to the +11JL image input terminal (+). In Fig. 7, the horizontal line above the symbol indicates that motion compensation has been performed, and A is the interpolation function [-8 due to 8
-+.

B represents the interpolated value interpolated from b+, CI, +It, and B represents the interpolated value interpolated from b+, dl by the interpolation function [B.

[Departure! ! Effect 11] As described above, according to the present invention, one field's worth of motion information is obtained by approximation from the transmitted motion information, and motion correction is performed for each field. Even if panning occurs, interpolation by the interfield interpolation filter can be stabilized, and interpolation can be performed with less error, thereby suppressing a decrease in resolution during panning.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an example of a schematic configuration of a motion compensation subsample interpolation method according to an embodiment of the present invention, and FIG. 2 shows an example of a motion vector at the time of baninking to explain the operation of FIG. 1. 3 is a diagram showing the existence range of inter-field motion vectors separated by 1 frame and inter-field motion vectors separated by 1 field. FIGS. 4, 5, and 6 are vector diagrams showing fields separated by 1 frame, respectively. FIG. 7 is a timing chart showing the flow of signal t'/ in the motion compensation sub-sample interpolation method according to the present invention. FIG. Turok IA shows an example of the schematic configuration of the conventional motion compensation 11-miss sample interpolation method.
Figure 9 is a timing chart showing the flow of motion when motion compensation IE is not performed using the conventional motion compensation subsample interpolation method, and Figure 10 is a motion capture IF using the conventional motion compensation subsample interpolation method. 3 is a timing chart showing the flow of signals when performing the following steps. (1)...Video input terminal t, (2)...Switch, (
3)...11j calculator, (4)...first motion correction field memory, (5)...second motion correction field memory. (6)...Interfield interpolation filter, (7)...
In-field filter, (8)...Switch, (
9)...Video output terminal, (lO)...Non-motion compensation field memory. In addition, in the figures, the same reference numerals indicate the same or corresponding parts.

Claims (1)

[Claims] (1) Based on the sample values that are intermittently sub-sampled and transmitted in a cycle of 4 fields while maintaining a predetermined sample position, the receiving side interpolates the missing sample values received during the 4-field period. This is a subsample interpolation method that reproduces a video signal by delaying and interpolating sample values on the receiving side based on motion information between fields separated by one frame of the video signal detected on the transmitting side. In the motion compensation sub-sample interpolation method that corrects the interpolation position, the receiving side uses motion information detected on the transmitting side and the motion information between the field of interest and a past field that is one frame apart. The motion information of the past field separated by one field from the field is approximately predicted to be 1/2 of the motion information on the sending side, and the interpolation position of the sample values of the past three fields is calculated from only the predicted motion information. 1. A motion compensation subsample interpolation method, characterized in that the motion compensation subsample interpolation method is configured to perform inter-field interpolation by correcting.