JPH074024B2 - Motion compensation type luminance signal / color signal separation circuit - Google Patents

Description

The present invention relates to a motion compensation type luminance signal / color signal separation circuit,
In particular, the present invention relates to a YC separation circuit in the time direction such as between frames or between fields using the motion vector between frames and fields.

2. Description of the Related Art In recent years, many methods have been proposed for accurately receiving current television broadcasts and transmitting high-definition signals while maintaining compatibility. Among them, as a method for improving the vertical resolution, a method of converting a conventional interlaced scan into a non-interlaced (sequential) scan by using a memory is considered. To improve the horizontal resolution, the high frequency component of the luminance signal is stacked on the color signal and transmitted (see, for example, Japanese Examined Patent Publication 60).
-12883), a method of multiplexing in a three-dimensional frequency gap (for example, Japanese Patent Publication No. 59-171387) can be considered. In the YC separation circuit, in addition to the conventionally used horizontal band pass filter (hereinafter abbreviated as BPF), a vertical axis filter (so-called comb) that uses a carrier color signal that is 180 ° inversion of the phase before one scanning line is used. Shape filter)
Alternatively, a method of using a scanning line in the vicinity of (1) field before or a scanning line in (1) frame before is considered. In addition, in order to avoid deterioration due to moving images, which is a drawback when using fields and frames, a motion adaptive type in which the above methods are switched and used is also proposed. (For example, Japanese Patent Publication Sho 58-129892
An example of the motion adaptive YC separation circuit will be described below with reference to the drawings.

4 and 5 show examples of block diagrams of conventional motion adaptive YC separation circuits. In FIG. 4, 1 is a field memory, 2 is 1H (H indicates a horizontal scanning period) memory, 3 is a switching circuit, and two multipliers 31 and 32 and an adder 33 are provided.
It is composed of. 4, 7 are subtractors, 5 are 1/2 coefficient circuits, 6
Is a BPF, and 8 is a motion detector. On the other hand, FIG. 5 shows another example, in which 11 frame memories are replaced with the field memories shown in FIG. In FIGS. 4 and 5, the input composite video signal is introduced into the field memory and 1H memory or the frame memory and 1H memory and switched by the output of the motion detector. Fig. 4 is for switching the color separation circuit using the field memory to the color separation circuit using the 1H memory, and Fig. 5 is the color separation circuit using the frame memory for the color separation circuit using the 1H memory. It is something to switch. The coefficients k ₁ and k ₂ in the figure may be 4 bits, for example, and they are complementary, and k ₁ becomes smaller and k ₂ becomes larger in accordance with the amount of motion. The signal switched by the switching circuit 13 or 3 is subtracted from the original composite video signal, and the 1/2 coefficient circuit and B
The color signal C is obtained through PF. On the other hand, for the luminance signal Y, the color signal C thus obtained is compared with the original composite video signal by 7 or 17
It is obtained by subtracting with the subtractor of.

By setting both the coefficients k ₁ and k ₂ which are the outputs of the motion detector 8 or 18 to 0, the output of the switching circuit can be set to 0. In this case, since the color signal C is separated by the BPF, it constitutes a horizontal axis YC separation circuit.

FIG. 6 is a detailed block diagram of an example of the motion detector 8 or 18, which is shown including three field memories. There are three field memories, 21-23, 21 and 22 or
22 and 23 correspond to the frame memory. Below these field memories, a motion detector based mainly on color signals is used to obtain the frame difference of the field difference. Since the field memories 21 and 23 are 262H, the phases of the carrier waves included in the signals at both ends thereof are the same. Therefore, if the difference is taken, a signal corresponding to the amount of movement of the color signal component can be obtained. However, since the positions of the corresponding scanning lines are displaced between the fields, the true amount of motion can be detected by comparing with the difference between the fields one frame before. In FIG. 6, 24 and 25 are inter-field subtractors, 26 and 27 are BPFs for extracting color signal components, 28 and 29 are absolute value circuits, 30 is an inter-frame subtractor and 31 is an absolute value circuit. . On the other hand, above the field memories 21 to 23, the field detector of the frame difference is mainly obtained by the motion detector corresponding to the luminance signal. Since the phase of the carrier wave is the same between frames, a signal corresponding to the amount of movement of the luminance signal component can be obtained by taking the difference. However, even if there is no movement between frames, there may be movement between fields, so the true amount of movement can be detected by adding the difference between frames of the preceding and following fields.

In FIG. 6, 32 and 33 are interframe subtractors, 34 and 35 are LPFs for extracting a luminance signal, 36 and 37 are absolute value circuits, 38
Is an inter-field adder. The total motion detection is obtained by adding the frame difference of the inter-field difference corresponding to the color signal component and the field sum of the inter-frame difference corresponding to the luminance signal component.

The conventional motion adaptive YC separation circuit is a method of switching the frame-use YC separation or the field-use YC separation to the line-use YC separation according to the amount of movement of each pixel in the screen. That is, in a portion with a large amount of movement, YC separation using 1H memory is selected, and diagonal cross color and dot interference cannot be removed.

Problems to be Solved by the Invention Since YC separation using 1H memory is spatially based on the information of one horizontal scanning line before, color is changed in a portion where the color greatly changes between lines in the vertical direction. There was a drawback of deterioration. On the other hand, YC separation using fields and frames does not work effectively on moving parts, but rather causes deterioration.
The motion-adaptive YC separation circuit shown in the conventional example determines the amount of motion for each pixel, and uses YC separation using frames (fields) for areas with little motion or static, and uses lines for areas with large motion. It was to selectively control YC separation and horizontal axis separation. When observing generally broadcast images, it is often seen that an image close to a still image moves slowly due to panning (horizontal movement) or tilt (vertical movement) of the camera. On the other hand, since the human eye can track a relatively slow movement, a difference between a still image and a moving image is often noticeable. That is, in the case of a still image, the frame-based YC separation circuit is selected, so a good image without cross color or dot interference can be obtained, but once it starts moving, it switches to line-based YC separation, so the interference is sudden. It will stand out.

As described above, simply switching the frame-based YC separation and the field-based YC separation to the line-based YC separation simply by detecting the motion amount does not mean that the respective YC separation circuits are fully utilized, and the slow motion In this case, there was a drawback that the difference was noticeable and interference appeared.

In the present invention, in order to solve the above problems, by selecting frame-based YC separation or field-based YC separation without deterioration even when the camera moves slowly such as panning, or when a small object moves, positive crossing is performed. It is intended to provide a circuit capable of removing color and dot interference.

Means for Solving the Problems In order to solve the above problems, the YC separation circuit of the present invention has a circuit that divides the entire screen into several blocks and obtains a motion vector between fields or frames. The present invention is characterized in that means for shifting the previous field or previous frame to be used according to the motion vector and superposing it on the current field or frame is used, and the YC separation using frame or field is used without deterioration.

Action The typical momentum of the object contained in a block is known by dividing the entire screen into blocks and obtaining the motion vector. While moving the previous field or frame according to this motion vector, the amount of motion is detected by the conventional motion detection circuit for each pixel in the current field or frame,
Depending on the combination, still areas use YC separation using frames or fields, moving areas use YC separation using moved frames or fields, and moving areas use YC separation using lines in areas without corresponding moving frames or fields. Each can be selected. As a result, the field or frame-based YC separation will be used in more areas than before, and the field or frame-based YC separation will be applied to small objects that are particularly noticeable due to slow motion or for small objects that move within a block. Therefore, a good image without cross color or dot interference can be obtained.

Embodiment A YC separation circuit according to an embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows a block diagram of a YC separation circuit in the first embodiment of the present invention. In FIG. 1, 1 is a frame memory, 2 is a 1H memory, 3 is a switching circuit, 4 is an adder, 5 is a 1/2 coefficient circuit, 6 is a band pass filter (B
PF), 7 is a subtractor, 8 is a motion detector, 9 is a motion vector detector, and 11 is a switch. 1 frame memory is 4
The frame use YC separation circuit is configured together with the adder / subtractor of 10 and the switcher of 10 and the 1H memory of 2 similarly configures the line use YC separation circuit. The switching circuit of 3 is the output of the motion detector of 8.
The frame YC separation and the line YC separation are weighted and switched by k ₁ and k ₂ . The output of the adder / subtractor 4 is output as the color signal C through the 1/2 coefficient circuit of 5 and the BPF of 6. On the other hand, the luminance signal Y is obtained by subtracting the color signal C from the original composite video signal by a subtractor of 7. The motion vector detector 9 is the main means of the present invention as described above, and detects the inter-frame motion vector for each block, and reads the frame memory 1 accordingly and controls the address to detect the previous frame. The coefficients k ₁ and k ₂ are controlled so as to select the frame-based YC separation by moving and reading only the motion vector thus obtained and superimposing it on the current frame. In addition, the phase of the subcarrier of the current frame and that of the previous frame that has moved differs depending on the motion vector.
It is necessary to switch the input signal of 1 by the switch 10 to adjust.

FIG. 2 is a detailed block diagram of the motion vector detector, and FIG. 3 is an explanatory diagram of intra-region switching in the Nth block. The motion vector detection for each block is a technique developed and put to practical use in motion-compensated interframe predictive coding. Various algorithms have been proposed, but basically, a method called a matching method is used to evaluate the difference between both frames by considering all motion vectors and obtain an optimum vector. FIG. 2 shows an example of a circuit that realizes the method. In FIG. 2, a video signal is input and supplied to the frame memory 21 and the comparator 22. FIG. 2 shows the motion vector detector of the Nth block, and it is considered that the same detector is prepared for the number of blocks. It is sufficient for the 21 frame memories to have a built-in number of elements corresponding to the number of horizontal scanning lines and horizontal pixels included in the Nth block. The comparator 22 accumulates, for each pixel, the difference between the current frame and the previous frame moved by the motion vector by the controller 24 and the address controller 23. The controller sequentially generates all possible motion vectors and instructs the address controller 23 on the read start address. The address controller changes the position of the previous frame supplied to the comparator 22 by changing the read start address of the frame memory. The controller obtains the error signal corresponding to the output motion vector from the comparator, accumulates it, tries all the motion vectors, evaluates the error signal and detects and outputs the optimum motion vector.

FIG. 3 shows an example of the Nth block. It is assumed that the motion vector is detected by the motion vector detector as shown in FIG. 2, and it is shown in the figure. On the other hand, the motion detector detects a still area C and a motion area A + A '+ B'. Here, A indicates the object position of the previous frame, and A ′ indicates the object position of the current frame. B'is an object that has flowed in from the outside of the block in the current frame. A feature of the present invention is that YC separation of the object A ′ portion in the current frame is performed using the previous frame A. Motion adaptive YC using conventional motion detector
In the separation circuit, the object A'section was performed by line-based YC separation or horizontal axis separation. In the present invention, frame-based YC separation can be used in the A region by moving and superimposing the previous frame as shown by the dotted line in FIG. 3 based on the motion vector detected by the motion vector detector. On the other hand, for the motion areas A and B, line-based YC separation is used as usual.

As described above, according to the present invention, by detecting the motion vector between fields or frames for each block, the previous field or frame is moved according to the motion vector to add or subtract with the current field or frame to obtain the color signal. Can be separated. Further, as in the conventional case, the field or frame YC separation that does not move can be applied to the still area, and the line YC separation can be used in the moving area where there is no reference field or frame. Thus, while line-based YC separation or horizontal axis separation was used in the entire motion area in the past, frame or field-based YC separation can be used in more parts by using motion vectors, and cross color and It is possible to realize a YC separation circuit with little dot interference. The improvement effect is more remarkable for slow movements such as panning of the camera.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a detailed block diagram of an example of a motion vector detector used therein, and FIG. 3 is an N-th diagram for explaining the embodiment of FIG. FIGS. 4 and 5 are block diagrams of a conventional motion adaptive YC separation circuit, and FIG. 6 is a motion detection circuit diagram used for them as an example. 1 ... Field memory, 3 ... switching circuit, 4 ... adder, 8 ... motion detector, 9 ... motion vector detector, 10 ... switcher.

 ─────────────────────────────────────────────────── ─── Continued Front Page (72) Hideshi Inoue 1006 Kadoma, Kadoma City, Osaka Prefecture Matsushita Electric Industrial Co., Ltd. Industry Co., Ltd.

Claims (2)

[Claims] 1. A circuit for separating a carrier color signal from a composite color television signal in which the carrier color signal is superimposed on a luminance signal, has means for detecting a motion vector between fields or frames for each block, and the motion A motion compensation type luminance signal / chrominance signal separation circuit, characterized in that a carrier color signal and a luminance signal are separated by moving a previous field or frame according to a vector and adding or subtracting the current field or frame. 2. A luminance signal / chrominance signal which comprises a luminance signal / chrominance signal separation circuit using a line memory and which moves a previous field or frame according to the motion vector and adds or subtracts from the current field or frame. The motion compensation type luminance signal / color signal separation circuit according to claim (1), characterized in that the separation circuit is switched for each region or the mixing ratio is changed.