JPS61206384A - High picture quality device of television signal - Google Patents

Abstract

PURPOSE:To prevent a deterioration of a picture quality in a boundary of a block by dividing a picture plane into plural blocks, disposing a means for selecting a signal processing of plural modes corresponding to a movement every block, and using a mode information of the block and the mode information of the block adjacent thereto to perform the signal processing. CONSTITUTION:In a weight circuit 5, with respect to picture elements (a), (b), (c), (d), a weight 1 is applied to an output of an auxiliary filter 1 and a weight 0 is applied to an output of an interpolating filter except therefor. Accordingly, the output of the weight circuit 5 obtains a signal corresponding to (A+B)/2 interpolated by the auxiliary filter 1. For the picture element (e), since interpolation is done by an average of a mode M1 and a mode M2, in the weight circuit 5, a signal is obtained by applying a weight of 1/2 to outputs of an interpolating filter 1, and an interpolating filter 2. Namely, corresponding to the mode information, in the weight circuit, the output signal of the interpolating filter is weighed and thereby an average signal of the mode information can be produced.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to an apparatus for improving the image quality of television signals, and particularly to an apparatus for improving image quality suitable for reproducing high definition or high quality images.

[Background of the invention]

High-definition televisions, or high-definition televisions, transmit and receive images that are several times more precise than current televisions through signal processing on the image sending and receiving sides. signal processing.

A so-called motion adaptive type is used. Namely.

It has parameters for a plurality of modes, and performs signal processing by switching to the optimal mode parameters for the movement of an image such as a still image or a moving image. As shown in Figure 1, these motion adaptive processes often divide the screen into multiple blocks and select the optimal mode for each block to perform signal processing (one-channel satellite transmission for high-definition television). Method (MUSE), TV Technical Report T E B S
95-2 (March 1982), on the image transmitting side, this mode information is also multiplexed with the signal and transmitted, and on the receiving side, signal processing is performed using mode information corresponding to each block. Therefore, when adjacent blocks are in different modes, a problem arises in that the image quality changes at the boundary between the blocks on the reproduced image.

That is, in adaptive processing that simply follows mode information.

This has the problem of deterioration of image quality at block boundaries, etc.

[Purpose of the invention]

SUMMARY OF THE INVENTION An object of the present invention is to provide an apparatus for improving the image quality of a television signal with less deterioration in image quality even when adjacent blocks undergo motion adaptive processing in different modes.

[Summary of the invention]

The present invention aims to achieve the above object by reproducing an image at a block boundary in a mode that is an average of the modes of adjacent blocks, and this is shown in FIG.

In FIG. 2, the mode information of blocks in the i-th and j-th columns is indicated by M t,j. In addition, mode information M i, J
The signal is processed and the pixel is denoted by f (M, , , )1
In the present invention, at the boundary between the block i-1,j and the block 1wJ, mode information (Mi-1,,1+M(,J
)/2 signal processed pixel f ((Mt-i, J +
Ml, J)/2) is used. In addition, at the boundary with i, j−′l, mode information (M = , J−0”Mi,
, +)/2.

Furthermore, mode information (M th-x ,
j+ M l , J + M t , J-□)/3
Perform signal processing.

For the remaining pixels, signal processing is performed using mode information M, .

Hereinafter, similar processing is performed for each block.

As described above, in the present invention, the mode information is averaged at the boundaries of blocks, so the joints between blocks become smooth, and image quality deterioration does not occur at the boundaries.

Note that mode averaging at block boundaries.

Various methods other than those shown in FIG. 2 can be considered. For example, as shown in FIG. 3, it is also possible to give a gradient characteristic to the weighting ratio of mode information for a plurality of pixels at the boundary. .

[Embodiments of the invention]

An embodiment of the present invention will be described below.

FIG. 4 shows a case where the motion adaptive signal processing of the present invention is applied to signal processing for interpolation of pixels on the receiving side.

This example shows a case where there are four types of mode information. Interpolation filter 1 has mode information Ml, and interpolation filter 2 has mode information M1. In interpolation filter 3, mode information M
2. The interpolation filter 4 performs an operation corresponding to the mode information M4. The signals obtained by these interpolation filters are inputted to a weighted circuit 5, and mode information is also inputted to the weighted circuit 5.

According to this mode information, the output signals of the interpolation filters corresponding to each mode are also weighted.

Now, pixels a and b span between two blocks.

Consider Q+ de Be fy ge h. Among these, a to d are mode information M1. It is assumed that eh is mode information M2. Furthermore, for pixel interpolation, scanning line doubling interpolation as shown in FIG. 5 is considered, and interpolation filter 1. Let us consider the characteristics of interpolation filter 2.

Here, it is assumed that pixels a, b, c, and d are interpolated in modes M1 and e.

In circuit 5, weighting is performed by weighting 1 to the output of the interpolation filter for pixels a, b, c, and d, and weighting O to the outputs of the other interpolation filters.Therefore, weighting is performed by circuit 5. A signal corresponding to -(A+B) interpolated by the interpolation filter is obtained as the output of In circuit 5, the outputs of 2 are weighted by 1/2, respectively. The output of the interpolation filter is -(A+B), and the output of the interpolation filter 2 is -(A+B+C+D). Further, since one pixel fvgth is interpolated in mode M2, the weighting circuit 5 gives a weight of 1 to the output of the interpolation filter 2 and a weight of 0 to the other interpolation filters. So in this case.

The signal of interpolation filter 2 -(A+B+C+D) is obtained.

As described above, by weighting the output signal of the interpolation filter using a weighting circuit in accordance with the mode information, it is possible to generate an averaged signal of the mode information. In Figure 6.

This weighting shows the detailed configuration of the circuit 5. multiplication circuit 6,
7, 8.9, weighting is realized by which multiplication is performed on the output signal and coefficients of the interpolation filter.

For these coefficients, a coefficient generation circuit 10 generates coefficient values corresponding to the mode information. It is also possible to configure this with a ROM, for example. And addition port 11
As shown in Figure 7, the mode information is, for example, information on the amplitude level corresponding to each mode information, or as a code, on the first line of each field. Multiplex.

As described above, according to this embodiment, image quality deterioration at boundaries between blocks can be removed with a simple configuration.

On the other hand, the mode information of each block often changes over time.For example, if a specific block changes alternately between mode information M1 and mode information M2 over time, the image quality will vary depending on the mode. Since the values are different, the blocks may appear to flicker visually. In order to eliminate this, signal processing may be performed using the average of the mode information in the time direction as well. This embodiment is shown in FIG. This embodiment also deals with pixel interpolation as signal processing. The configuration is the same as that in FIG. 4 except for the frame delay circuit 16 for mode information.

The interpolation filter 12 includes mode information M1, interpolation filter 1
3 is mode information M2. The interpolation filter 14 uses mode information M
3. The interpolation filter 15 performs signal processing corresponding to the mode information M4. The weighting circuit 17 weights the output of each interpolation filter.

Here, it is assumed that in the previous frame, all mode information was M, and in the current frame, the mode information was M1 and M2. In this case, pixel a of the current frame.

b, Q, . . . h are each generated by the method shown below.

First, since the mode information of the previous frame and the current frame of pixels a, b, Q, and d match, they become signals of the mode information M1 of the interpolation filter 12, as shown in the previous embodiment. Regarding the pixel e, as shown in the previous embodiment, the interpolation filter 12. 6 On the other hand, for pixels f2g and h, since the previous frame is Mo and the current frame is M2 mode information, the average mode of M1 and M2 is determined by the interpolation filter 13. take. Therefore, the signal is obtained by weighting the interpolation filter 12 and the interpolation filter 13 by 172.

Furthermore, since the configuration of the circuit 17 for weighting may be almost the same as that shown in FIG. 6, a description thereof will be omitted.

Note that although mode information of two frames is used in this embodiment, control may be performed using information of more frames.

〔Effect of the invention〕

According to the present invention, even when the mode information of adjacent blocks is different, motion adaptive processing that does not cause image quality deterioration at the boundary between blocks becomes possible, which is extremely effective in improving image quality.

Although the embodiment has been described using interpolation signal processing as an example, it goes without saying that the present invention can also be applied to motion adaptive processing such as YC separation on the receiving side.

[Brief explanation of drawings]

FIG. 1 is a conceptual diagram of motion adaptive processing divided into blocks, FIGS. 2 and 3 are diagrams explaining the principle of the present invention, and FIGS. 4 to 8 are examples of the present invention. 1.2, 3, 4, 12, 13, 14.15...Interpolation filter, 5.17...Weighting circuit, 16...Frame delay circuit, 6,7,8.9... Multiplying circuit, 10 eyes 2nd ○: child (Nipri■・chi (morning・1 shiji ○° child's astonishment Chiji■: f(♀・N,, '-t,, ward・child 3rd mouth ○ : f (Mt-t,;) ○: f (,:3 竺-1pi ” J/Yi, 诹) No. 4 Fig. 7 'l 74ny 2 6th exit J7121

Claims (1)

[Claims] It has means for dividing the screen into multiple blocks and selecting multiple modes of signal processing according to the movement of each block, and performs signal processing using the mode information of the above block and the mode information of adjacent blocks. An apparatus for improving the image quality of television signals, characterized by performing the following.