JPH07114496B2 - Motion compensation subsample interpolator - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of application] The present invention, in a high-definition television receiver, calculates a motion correction amount between each field from transmitted motion information, and interpolates a missing point between the fields. The present invention relates to a motion correction subsample interpolation device.

[Conventional technology]

The conventionally proposed motion-correction subsample interpolation device is
It is described in the material announced by NHK Broadcasting Technology Research Laboratories in "New Transmission System (MUSE) of High Definition Television" at the foundation commemorative lecture in June, 1984. According to this, a motion-correction sub-sample interpolation device having a configuration as shown in FIG. 4 is considered.

In the figure, 1 is a video input terminal for inputting a video signal transmitted at a sample rate of 16.2 MHz, and 2 is a video signal input to this input terminal 1 and an output signal of a motion compensation field memory 7 described later and 32.4 MHz. Switches S ₁ and 3 that switch at subsample timing pass through switch S ₁ .
A non-motion-correction field memory that stores a 1-frame signal having a sample rate of 32.4 MHz, 7 stores a 1-frame signal that has a sample rate of 32.4 MHz output from this memory 3, and performs motion correction using a motion vector. This is a motion compensation field memory to be executed. 11 missing on the basis of the signal passed through the output signal and the switch S ₁ non-motion compensated field memory 3, the inter-field interpolation filter for interpolating the missing sampling points, 12 on the basis of only the signal passing through the switch S ₁ In-field interpolation filter to interpolate sample points, 13
Is connected to the upper contact when no motion correction is performed to allow the output signal of the inter-field interpolation filter 11 to pass through, and is connected to the lower contact when motion correction is performed or when motion detection is performed and Switches S ₂ and 14 for passing the output signal are signals passing through the switch S ₂ ,
That is, it is a video output terminal for outputting a signal having a sample rate of 64.8 MHz by interpolating the missing sample points.

Next, the operation will be described. A new transmission method for high-definition television signals is sub-Nyquist sampling, which makes a cycle of four fields, and its required bandwidth is 8.1 MHz.

Now, consider the case where the signal of the first field is input to the video input terminal 1. At this time, the non-motion correction field memory 3 stores the signals of the second field and the fourth field before one cycle, while the motion correction field memory 7 stores the signals of the first field and the third field before one cycle. ing. When a motion vector exists due to panning of the camera, the contents of the motion correction field memory 7 move two-dimensionally according to the vector amount. Here, it is assumed that the transmitted motion vector is a motion correction amount between frames separated by one frame. On the other hand, the motion correction amount to be input to the motion correction field memory 7 is the motion correction amount between fields separated by one frame, but considering that the motion of the image by panning is sufficiently smooth,
It can be considered that the contents of the motion correction field memory 7 are subjected to motion correction based on the signal of the first field input to the video input terminal 1. The switch S ₁ is switched at the timing of the sub-sample of 32.4 MHz, the phase is inverted for each field, and the phase is also inverted by the motion vector. After all, in the present case, the first input from the video input terminal 1
Field signal and motion compensation was done 3rd before one cycle
The field signal passes through switch S ₁ .

The signal that has passed through the switch S ₁ and the output signal of the non-motion correction field memory 3 are input to the inter-field interpolation filter 11 to perform inter-field interpolation. When motion compensation is not performed, that is, when switch S ₂ is connected to the upper contact, inter-field interpolation of 64.8 MH from video output terminal 14 is performed.
Outputs a video signal with a sample rate of z.

The signal that has passed through the switch S ₁ is input to the intra-field interpolation filter 12. The intra-field interpolation filter 12 performs intra-field interpolation, and when the switch S ₂ is connected to the lower contact, outputs the video signal at the sample rate of 64.8 MHz interpolated in the field from the video output terminal 14. The switch S ₂ is connected to the lower contact is when the time and the motion detection is made that motion compensation is performed, the former field by field, the latter switch S ₂ is switched in units of pixels.

FIG. 5 shows the states of 5a to 5e in FIG. 4 when the C ₁ field is input to the video input terminal 1 without motion compensation. However, if the signal is expressed in fields ... a ₀ , b ₀ , c ₀ , d
It is assumed that the flow is ₀ , a ₁ , b ₁ , c ₁ , d ₁ ... A in the figure
D ₀ by the interpolation function f _A is the inter-field processing, a _1, b _1, c
Indicates the interpolation value that is interpolated from ₁ . FIG. 6 shows the states of the signals 6a, 6c to 6e in FIG. 4 when the motion correction is performed and the c ₁ field is input to the video input terminal 1. In the figure,
The horizontal line above the signal indicates that motion compensation has been performed, B
Is _{▲ ▼} c ₁ due to the interpolation relation f _B
The interpolation value interpolated from is shown.

[Problems to be solved by the invention]

In the conventional motion compensation sub-sample interpolation method, when the motion compensation is performed, the transmitted vector is used as a motion compensation amount between the fields separated by two fields, and the motion compensation between adjacent fields is not performed. Therefore, the interpolation of the missing sample points has to be performed in the field, and this is interpolated by the in-field interpolation filter. However, this method has a problem that the resolution at the time of performing the motion correction is lowered.

The present invention has been made to solve the problems as described above. The motion correction amount between each field is accurately obtained from the transmitted motion vector, and the motion correction is performed similarly to the case where the motion correction is not performed. An object of the present invention is to obtain a motion-compensated sub-sample interpolation device capable of suppressing a decrease in resolution by using an inter-field interpolation filter even when performing correction.

[Means for solving problems]

A motion compensation sub-sample interpolation apparatus according to the present invention, on the basis of a sample value of a video signal transmitted by intermittently performing sub-sampling so as to make a cycle of four fields while maintaining a predetermined sample position. When the video signal is reproduced by interpolating the missing points from the sample values received in the 4 field period, the receiving side is based on the motion information between the frames of the video signal detected by the transmitting side one frame apart. In a motion correction sub-sample interpolation device that delays a video signal and corrects the interpolation position of a sample value, motion information between fields separated by one frame on the receiving side is calculated from motion information between frames separated by one frame. A first motion correction information calculation means to be obtained, and the first
First motion correction means for correcting the motion of the video information separated by one frame based on the output of the motion correction information calculation means of
Second motion information calculation means for obtaining motion information between fields separated by one field from the first motion correction information calculation means, and video information separated by one field based on the output of the second motion correction information calculation means. Using the second motion compensation means for compensating the motion of the present field, the motion information between the fields separated by one frame and the motion information between the fields separated by one field, which are obtained on the receiving side, with respect to the current field. And the inter-field interpolating means for interpolating the sample points missing during transmission from the motion-compensated video signals of a total of four fields are inter-field interpolated. is there.

[Action]

According to the present invention, with the above-described configuration, the motion vector between fields separated by one frame and the motion vector between fields separated by one field can be accurately obtained from the transmitted motion vector. However, in the conventional motion compensation, the sample points that are missing during transmission are interpolated by the intra-field interpolation, and as a result, the resolution of the image to be motion compensated is lowered, whereas in the motion compensation of the present invention, , The video signals of the past three fields can be motion-corrected with respect to the current field by the above-mentioned two types of motion vectors obtained by the receiving side. Inter-field interpolation from the video signal makes it possible to interpolate sample points that are missing during transmission, resulting in motion It is possible to suppress deterioration in resolution of an image Tadashisa.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings. First
In the figure, 1 is a video input terminal for inputting a video signal transmitted at a sample rate of 16.2 MHz, 2 is the input terminal 1
The first motion compensation field switch S ₁ for switching between the output signal of the memory 7 at the timing of the sub-sample of 32.4 MHz, 3 is a sample rate of 32.4 MHz to pass through the switch S ₁ to be described later with the video signal inputted to the 1 It is a non-motion correction field memory that stores signals for frames. 4 is a multiplier for multiplying the transmitted motion vector by 2, 5 is a multiplier 4
From the value of twice the motion vector which is the output of
1 is a subtracter for reducing the output of the 1-field delay device 6, and 6 is a first 1-field delay device for delaying the output signal of the subtractor 5 by 1 field.
A first motion correction information calculation means for obtaining a motion vector between fields separated by one frame from the transmitted motion vector between frames is configured. Reference numeral 7 is a first motion compensation field memory which stores a signal for one frame having a sample rate of 32.4 MHz output from the non-motion compensation field memory 3 and performs motion compensation by the output signal of the subtractor 5. Together with the field memory 3 and the switch S1, a first motion correction means for performing the first motion correction based on the output of the first motion correction information calculation means is configured. Further, 8 is a subtractor for subtracting the output of a second 1-field delay device 9 described later from the output signal of the subtractor 5, and 9 is a second 1-field delay device for delaying the output of the subtractor 8 by 1 field, By these, the motion vector between fields separated by one field is obtained.
Motion compensation information calculation means. Reference numeral 10 is a second motion correction field memory which stores a signal for one frame having a sample rate of 16.2 MHz passing through the switch S ₁ and performs motion correction by the output signal of the subtractor 8. This second motion correction field memory The memory 10 constitutes second motion correction means for performing the second motion correction based on the output of the second motion correction information calculation means.

11 is an inter-field interpolation filter that interpolates a missing sample point based on the signal passing through the switch S ₁ and the output signal of the second motion compensation field memory 10, and 12 is based on only the signal passing through the switch S _1. In-field interpolation filter that interpolates the missing sample points, 13 is normally connected to the upper contact to pass the output signal of the inter-field interpolation filter 11 and connected to the lower contact in pixel units when the motion is detected switch S _2, 14 for passing the output signal of the inner interpolation filter 12 is a video output terminal signal passing through the switch S _2, i.e. missing sampling points to output a signal a sample rate of 64.8MHz is interpolated.

Next, the operation will be described. FIG. 2 shows motion vectors when panning occurs from the still image state in order to specifically explain the operation of FIG. In the figure, x represents the screen horizontal axis and y represents the screen vertical axis. Each field of the video signal is represented by c _{0 to} d ₂ (when the signal is represented by fields ... a ₀ , b ₀ , c ₀ , d ₀ , a ₁ , b ₁ , c ₁ , d ₁ , a ₂ , b ₂ , c ₂ , d ₂
......), the motion vector is ₀ ,
It is represented by ₁ , ₁ , ₁ , ₁ , ₁ , ₂ . The motion vector here means that it is a motion correction amount between frames separated by two fields. In order to further explain the definition of this motion vector, the amount of motion correction between fields separated by two fields is additionally expressed by ₁ , ₁ , ₁ , ₁ , ₁ , ₁ , ₂ , ₂ and the motion correction between fields separated by one field is also shown. The quantity is represented by ₁ , ₁ , ₁ , ₁ , ₁ , ₂ , ₂ , ₂ , ₂ . For example, ₁ is the amount of correction for motion compensation of the a ₁ field based on the c ₁ field, and is based on the b ₁ field.
a This is the amount of correction for motion compensation of _one field. Similarly, the motion vector α ₁ can be said to be a correction amount for performing motion correction on the basis of one frame composed of a ₁ field and b ₁ field with respect to 1 frame composed of c ₁ field and d ₁ field. _{Between 0} , ₀ , ₁ ... And ₀ , ₀ , ₁ ... , And ₀ , ₀ , ₁ ... And ₀ , ₀ , ₁
… Between The relationship is established. In the example shown in FIG. 2, the c ₀ field, the d ₀ field, and the a ₁ field are still images, and b
It means that panning is occurring after the _first field. Therefore ₀ = ₀ = 0 and therefore And then Becomes On the contrary, if ₀ , ₁ , ₁ , ₁ ... Is expressed as ₀ , ₀ , ₁ ... Further represented _{1, 1, 1} ... _{0, 1, 1} ..., become that way. In this way, if the initial state is a still image, the amount of motion correction between fields separated by two fields can be sequentially obtained from the transmitted motion vector, and 1 from the amount of motion correction between fields separated by two fields.
The amount of motion correction between fields separated by fields can be sequentially obtained. By using such an array of two types of motion correction amounts, it is possible to perform motion correction of the video signals for the past three fields with respect to the current field and interpolate the missing sample points from the video signals for these four fields. This is the essence of this invention.

Next, the signal flow will be described with reference to FIG. First, when a signal of the c ₁ field is input to the video input terminal 1, the b ₁ field, d
_{The 0} field signal is stored in the first motion compensation field memory 7 as the a ₁ field and c ₀ field signals. The motion vector transmitted at this time is δ _0, which is doubled by the multiplier 4 and the first 1 by the subtractor 5.
The motion correction amount ₀ between the fields separated by 2 fields before the output of the field delay unit 6 is subtracted, and the motion correction amount ₁ between fields separated by 2 fields is reduced.
To get In the first motion correction field memory 7, the signals of the a ₁ field and c ₀ field stored by the motion correction amount ₁ between the fields separated by 2 fields are two-dimensionally moved and input to the video input terminal. Motion correction is performed based on _one field. A ₁ thereby signal passing through the switch S ₁ is to c ₁ field and motion compensation is performed
It is a signal with the field.

On the other hand, the second motion correction field memory 10 stores signals of the b ₁ field and d ₀ field as in the non-motion correction field memory 3. At this time, the output of the subtracter 5 is a motion correction amount ₁ between the fields separated by 2 fields.
Then, the subtracter 8 subtracts the motion correction amount ₁ between the fields separated by 1 field before the output of the second 1-field delay unit 9 to obtain the motion correction amount ₁ between fields separated by 1 field. In the second motion correction field memory 10, the signals of the b ₁ field and the d ₀ field stored by the motion correction amount ₁ between the fields separated by 1 field move two-dimensionally and move based on the c ₁ field. Correction is performed. Thus d ₀ field motion compensation has been performed based on c ₁ field, a ₁ field, since the four fields of the signal b ₁ field and c ₁ field itself enters the inter-field interpolation filter 11 between fields interpolation It will be possible.

In this way, the first motion correction field memory 7 stores signals two fields before and four fields before the field input from the video input terminal 1, and the second motion correction field memory 10 stores the signals. Stores the signals of one field before and three fields before for the field input from the video input terminal 1. Therefore, two fields are separated from the motion vector which is the motion correction amount between the frames separated by two fields. A motion correction amount between fields and a motion correction amount between fields separated by one field must be obtained. In the above description, two types of motion correction amounts are stably and sequentially obtained when panning starts from the still image state, but similarly, the two types of motion correction amounts are stably and sequentially obtained when panning ends and the state returns to the still image state. Therefore, inter-field interpolation can be performed even when motion correction is performed as in the still image state. It should be noted that, as for a general moving image signal, interpolation of missing sample points cannot be performed by inter-field interpolation, so when the motion is detected, the switch S ₂
Is connected to the lower contact on a pixel-by-pixel basis, and the intra-field interpolation filter 12 performs inter-field interpolation.

The signal in which the missing sample points are interpolated by the inter-field interpolation filter 11 or the intra-field interpolation filter 12 has a sample rate of 64.8 MHz and is output from the video output terminal 14.

FIG. 3 shows the states of the signals 3a to 3g in FIG. 1 when the signal of the c ₁ field is input to the video input terminal 1. Third
In the figure, the horizontal line above the signal indicates that motion compensation has been performed.
Is the interpolation function f _{A ▲ ▼} , _{▲ ▼} , _{▲ ▼} , c ₁
, B is the interpolation value interpolated from _{▲ ▼} , c ₁ by the interpolation relation f _B.

〔The invention's effect〕

As described above, according to the motion compensation sub-sample interpolation apparatus of the present invention, the sample of the video signal transmitted by intermittently sub-sampling so as to make a cycle of four fields while maintaining a predetermined sample position. Based on the value, when the video signal is reproduced by interpolating the missing points from the sample value received in the four-field period on the reception side, the motion between the frames of the video signal detected by the transmission side separated by one frame. In a motion correction sub-sample interpolation device that corrects an interpolation position of a sample value that delays and interpolates a video signal on the basis of information, a motion compensation sub-sample interpolator separates one frame on the reception side from the motion information between frames separated by the one frame. A first motion correction information calculating means for obtaining motion information between fields, and one frame based on the output of the first motion correction information calculating means First to perform motion compensation of the image information that has been
Based on the output of the second motion correction information calculation means, and the second motion information calculation means for obtaining the motion information between the fields separated by one field from the first motion correction information calculation means. Using the second motion compensation means for compensating the motion of the video information separated by one field, the motion information between the fields separated by the one frame and the motion information between the fields separated by one field, which are obtained by the receiving side. An inter-field interpolating means for inter-field interpolating the sample points missing during transmission from the motion-compensated video signals of the past three fields for the current field and interpolating the motion-compensated video signals of the total of four fields. As a result, from the transmitted motion vector, as a result, the video signals of the past three fields are compared with the video signal of the current field. It is possible to interpolate the missing sample points by using the video signals for a total of 4 fields by the inter-field interpolating means, and the so-called panning in which the entire camera slowly moves in the horizontal direction to take an image occurs, and relatively. Even if the decrease in resolution during motion compensation is easily perceived by the human eye, it will be replaced with the conventional inter-field interpolation processing during motion compensation under the condition that the amount of information is larger than before. Since inter-field interpolation processing by the inter-field interpolation filter can be performed in the same manner as in the still image state, there is an effect that it is possible to suppress a decrease in resolution during panning, for example.

[Brief description of drawings]

FIG. 1 is a block diagram showing an example of a schematic configuration of a motion correction sub-sample interpolation device according to an embodiment of the present invention, and FIG. 2 is a motion vector and interfield during panning for explaining the operation of FIG. FIG. 3 is a vector diagram showing an example of the relationship with the motion compensation amount of FIG. 3, FIG. 3 is a timing chart diagram showing the signal flow of the motion compensation sub-sample interpolation device according to one embodiment of the present invention, and FIG. FIG. 5 is a block diagram showing an example of a schematic configuration of a sub-sample interpolation device. FIG. 5 is a timing chart diagram showing a signal flow when motion compensation of a conventional sub-sample interpolation device is not performed. FIG. 6 is a timing chart showing the flow of signals when performing motion correction of the motion correction subsample interpolation device of FIG. 1 ...... video input terminal, 2 ...... switches S _1, 3 ...... non motion compensated field memory, 4 ...... multiplier, 5,8 ...... subtractor,
6 ... First field delay device, 7 ... First motion correction field memory, 9 ... Second one field delay device,
10 …… Second motion compensation field memory, 11 …… Inter-field interpolation filter, 12 …… In-field interpolation filter, 13 …… Switch S ₂ , 14 …… Video output terminal. The same reference numerals in the drawings indicate the same or corresponding parts.

Claims (1)

[Claims] 1. A receiving side performs 4 sampling based on a sample value of a video signal transmitted by intermittently performing sub-sampling so as to make a cycle of 4 fields while maintaining a predetermined sample position.
When the video signal is reproduced by interpolating the missing points from the sample value received in the field period, the video signal is received by the receiving side based on the motion information between the frames of the video signal detected by the transmitting side. In a motion correction sub-sample interpolation device for delaying and interpolating a sample value for interpolating, a motion information between fields separated by one frame on the receiving side is calculated from motion information between frames separated by one frame. A first motion correction information calculation means, a first motion correction means for correcting the motion of video information separated by one frame based on the output of the first motion correction information calculation means, and the first motion correction information calculation Second motion information calculation means for obtaining motion information between fields separated by one field from the means, and based on the output of the second motion correction information calculation means Using the second motion compensation means for compensating the motion of the video information separated by one field, and the motion information between the fields separated by one frame and the motion information calculated by the receiving side. And inter-field interpolation means for interpolating the sample points missing during transmission from the motion-compensated video signals of the past three fields to the current field and interpolating the sample points missing during transmission. A motion-compensating subsample interpolation apparatus comprising: