JPS634781A - Action signal detecting circuit in digital television receiver - Google Patents

Abstract

PURPOSE:To remove a false signal and to increase the detecting accuracy of an action signal by detecting the level of respective frame difference signal between picture elements to be noticed and the ambient elements, obtaining respective 'dynamic' and 'static' deciding signals, taking the decision by majority of these deciding signals and executing the final dynamic and static determination of the picture element to be noticed. CONSTITUTION:A deciding signal sampling circuit 8, in which the deciding signal concerning respective picture elements is given in the sequence from a dynamic and static deciding circuit 7, samples the deciding signal S1 of a picture element P1 to be noticed and respective deciding signals S2-S5 of ambient picture elements P2-P5, and given to a majority vote circuit 11 by matching the timing respectively with onescanning line delaying elements 9a and 9b and one-sample delaying elements 10a-10d. The majority vote circuit 11 takes the decision by the majority of inputted deciding signals S1-S5, and when the deciding signals of the 'H' level corresponding to the 'dynamic' are many, determines that the picture element P1 to be noticed is a moving image. On the other hand, when the deciding signals of the 'L' level corresponding to the 'static' are many, the circuit determines that the picture element P1 to be noticed is a static picture even if the deciding signal S1 of the picture element P1 to be noticed is the 'H' level.

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field) The present invention relates to a motion signal detection circuit in a digital television receiver, which obtains a signal for distinguishing between a still image and a moving image from a video signal.

(Prior art) Interlaced scanning has no large area flicker disturbance and
This method is used in current standard televisions as a band compression method for analog signals that can reduce the number of frames per second. However, at a near-sighted distance of 4H (H: screen height), interlace disturbances such as interline flickering and line crawling are large, and the transmitted information is not fully utilized. Therefore, high-quality digital television receivers employ means to remove these image disturbances through spatiotemporal signal processing on the receiving side (Television Society Technical Report TEBS 83-4
P19-P24).

First, television scanning is regarded as a two-dimensional or three-dimensional sampling system, and image quality deterioration is prevented by an interpolation filter that converts interlaced scanning into sequential scanning based on analysis of spectral distribution. A 2:l interlaced scan image becomes substantially equivalent to an interlaced scan image in which the number of scanning lines is increased by sequential scan conversion, and the image quality can be significantly improved. Such an interpolation filter uses a plurality of field memories to perform scan conversion based on adding together the current video signal and the video signal stored in the memory. At this time, if the video signals are still images, it is sufficient to simply add them together, but if the video signals are video signals, simply adding them together will cause a time lag, which is called double image interference. Image quality deterioration occurs. Therefore, video signals require different video signal processing than still images. For this reason, a digital television receiver using such an interpolation filter is provided with a motion signal detection circuit for distinguishing between moving images and still images.

First, in order to obtain a discrimination signal for discriminating between a still image and a moving image from a video signal, as shown in FIG. 3, the level difference of the luminance signal between frames is determined. That is, the luminance signal input from the input terminal 1 is stored in the field memory 2a,
2b and 2c, each field is stored in order. The subtraction absolute value 2''i3a outputs a difference signal between the luminance signal input taken from the input terminal 1 and the field memory 2b.

- On the other hand, the subtraction absolute value unit 3b outputs a difference signal between the field memory 2c and the field memory 2a. These difference signals are given to the maximum value detection circuit 4. The maximum value detection circuit 4 combines these difference signals by /n and outputs a frame difference signal.

That is, in the case of a still image, the luminance signal level between frames is almost the same, so no frame difference signal is detected. - On the other hand, in the case of a moving image, the level of the luminance signal changes between frames, so a frame difference signal is detected.

However, in reality, the level of the luminance signal may change due to the darkness of the video signal, noise, etc.
Even in the case of a still image, a frame difference signal may occur. Such a frame difference signal becomes a false motion signal and causes an error in determining whether a still image is a moving image or a moving image. If an error occurs in determining whether an image is a moving image or a moving image, a problem arises in that a video signal of a still image is subjected to signal processing for a moving image, or vice versa, resulting in significant deterioration of image quality.

In order to improve this problem, a motion signal detection circuit as shown in FIG. 4 has been proposed (see the aforementioned Television Society of Japan technical fJ7 report).

That is, the frame difference signal obtained by the frame difference signal detection circuit shown in FIG. 3 is given to the coring control circuit 21. - On the other hand, field memory 2 shown in FIG.
A frame sum signal is obtained from the outputs of a and 2C, and this sum signal is given to the differential absolute value circuit 22. The coring control circuit 21 has a threshold level controlled by the size of the edge of the image given from the differential absolute value circuit 22, so that the coring control circuit 21 eliminates noise and false frames without losing the frame difference signal of the moving image. Remove the difference signal.

Single-shot noises left behind by the coring control circuit 21 are removed by the isolated point removal circuit 23. Frame difference No. 43 outputted from the isolated point removal circuit 23 is given to a 3-sample low-pass filter 24.

Here, the frame difference signal is temporally expanded, and portions where no frame difference signal occurs in the moving image area are interpolated. The frame difference signal obtained in this way is given to the dynamic/static determination circuit 25. The motion/static determination circuit 25 compares the level of the frame difference signal with the appropriately determined thread level and the level of the frame difference signal, and if the frame difference signal is larger, it is determined to be a moving image, and if it is smaller, it is determined to be a still image. Outputs a motion signal.

However, since the coring control circuit 21 removes false motion signals by cutting off frame difference signals of one low level, there is a problem that low-level frame difference signals due to original motion are also removed. There is a point. Furthermore, since false motion signals are not necessarily isolated, the isolated point removal circuit 23 can be said to be less effective in removing false motion signals.

In detecting a motion signal from the difference in luminance signal level between frames in this way, it is difficult to remove only the false motion signal without substantially affecting the original motion signal.

(Object of the Invention) The present invention has been made in view of the above circumstances, and aims to improve the detection accuracy of motion signals by removing false motion signals without affecting the original motion signals. The purpose is to

(Structure of the Invention) In order to achieve the above object, the present invention has the following structure.

That is, the motion signal detection circuit in the digital television receiver according to the present invention includes a frame difference signal detection means for detecting a frame difference signal for each pixel by calculating the difference in luminance signals between frames; A JJI/static determination means that determines "motion J" or "stillness J" for each pixel by comparing its level with the J skewer bell, and a determination signal is provided from the motion/static determination means. judgment signal extracting means for extracting and outputting the judgment signals of the pixel of interest and a plurality of surrounding pixels, and determining whether the pixel of interest is "moving" or ``The special feature is that it is equipped with a majority voting means that outputs a motion signal according to the motion.
I'm letting go.

Next, the present invention will be explained in detail.

Since the level of the luminance signal fluctuates due to jitter, noise, etc. of the video signal, the level of the frame difference signal also fluctuates. Therefore, the determination signals output from the motion/static determination means include not only determination signals due to original motion but also determination signals due to false motion. On the other hand, the determination signal due to the original movement has a certain degree of continuity, but the determination signal due to the false movement is sporadic. Therefore, in determining whether a pixel of interest is moving or still, a majority vote is taken of the determination signals of each pixel, including the surrounding pixels. As a result, if there are many motion determination signals, the pixel of interest is determined to be a motion picture, and a motion signal is output from the majority decision means.

(Example) Hereinafter, the present invention will be described in detail based on an example shown in the drawings.

FIG. 1 is a block diagram schematically showing the configuration of a motion signal detection circuit in a digital television receiver according to an embodiment of the present invention.

The luminance signal input from the input terminal l is given to the frame difference signal detection circuit 6. The frame difference signal detection circuit 6 is
This circuit outputs a frame difference f3, which is a level difference in the luminance signal of each pixel between frames, for each pixel. Specifically, since the configuration described in FIG. 3 is adopted, the explanation here will be omitted.

The frame difference signal output from the frame difference signal detection circuit 6 is given to the work/quiet determination circuit 7.

The motion/static determination circuit 7 compares the level of the frame difference signal with a reference level appropriately determined in advance, and if the frame difference signal is larger, it issues a determination signal indicating that the pixel is a moving image (e.g. , '+1 J level), and if the frame difference signal is smaller, a determination signal indicating that the pixel is a still image (for example, 'LJ level) is output.

The determination signal output from the dynamic/static determination circuit 7 is transmitted through one scanning line delay elements 9a, 9b and one sample delay element 10a-
The judgment No. 13 obtained by +74 from 10d is given to the extraction circuit 8. This judgment signal extracting circuit 8 appropriately delays the judgment (g) given from the dynamic/static judgment circuit 7, and selects the pixel of interest Pi as shown in No. 212I,
The determination signals 31 to S5 of the surrounding pixels P2 to P5 are output in parallel. In the second section 1, L1 to L3 indicate scanning lines.

Judgment signal 31- outputted from judgment signal extraction circuit 8
35 is given to the majority circuit 11. The multi-dispersion circuit 11 determines whether the pixel of interest PI is "moving" or "static" by taking a majority vote of motion or stillness from the determination signal 5t-35, and outputs a motion signal corresponding thereto. This motion signal is applied to a 3-sample low-pass filter 12 similar to that described in FIG. 4 for interpolation. The interpolated motion signal is given to a motion signal processing section (not shown) and is used as a control signal for switching between still image and moving image processing methods.

Next, the operation of the embodiment having the above-described configuration will be explained.

Since the level of the brightness signal is affected by the jitter and noise of the video signal, even if a certain pixel is a still image, the frame difference signal detection circuit 6 may output a frame difference signal corresponding to a moving image. .

Therefore, the determination signal output from the motion/static determination circuit 7 is a mixture of determination signals due to the original motion and determination signals due to the false motion.

The judgment signal extracting circuit 8, which is sequentially given the judgment signal for each pixel from the dynamic/static judgment circuit 7, receives the judgment signal S1 of the pixel of interest P1 shown in FIG.
5, each determination signal 82 to S5 is extracted. That is, the determination signal S1 of the pixel of interest PI is output via the one-scan line delay element 9a and the one-sample delay element tOC. Also,
The determination signal S4 of the surrounding pixel P4 is transmitted through a one-sample delay element l.
Output via Oa. Judgment signal S5 of surrounding pixel P5
The determination signal S2 of the 0-period pixel P2 outputted via the l scanning line delay element 9a is outputted via the l scanning line delay element 9a, 9b
and 0 outputted through the 1-sample delay element lOb.
The determination signal S3 of the surrounding pixel P3 is one scanning line delay prevention 9a.
and is outputted via one sample delay blockers 10c and 10d.

In this way, the determination signal St of the pixel of interest PI and the determination signals S2 to S5 of the surrounding pixels P2 to P5 are transmitted to the l scanning line delay elements 9a, 9b and the one sample delay elements 10a to 10a.
10d, the signals are applied to the majority circuit 11 at the same timing.

By the way, since the movement of an object is continuous to some extent, 'i' is detected based on the movement of the object.
The JJ J determination signals are considered to be continuous to some extent. On the other hand, since the noise superimposed on the video signal is randomly generated and is one-off, the "motion" determination signal due to false movement detected based on such noise is one-off.

Therefore, the determination signal S1 of the pixel of interest PI is, for example, r
If there is a 'Hi level corresponding to 'motion' and this is due to the original movement, the surrounding pixels P2 to P5
Most of the determination signals 82 to S5 should be at the 'HJ level. Conversely, the determination signal S1 of the pixel of interest Pi is 'H
Although it is at the J level, if this is due to false movement, most of the judgment signals of the surrounding pixels P2 to P5 should be at the "LJ level" corresponding to "still". The majority circuit 11 takes a majority vote of the input determination signals 81 to S5, and determines that the target pixel PI is a moving image if there are many determination signals of 'HJ level corresponding to "motion". - On the other hand, if there are many determination signals at the 'LJ level corresponding to static J, even if the determination signal S1 of the pixel of interest P1 is at the 'HJ level, the pixel of interest P1 is determined to be a still image. Then, when it is determined that the pixel of interest PI is a moving image,
For example, the majority circuit 11 outputs a motion signal at the 'HJ level, and when it is determined that the image is a still image, outputs a motion signal at the 'LJ level. In this way, the motion/static determination for each pixel is performed in sequence.

The motion signal output from the majority circuit 11 is given to a motion processing section (not shown) via a 3-sample low-pass filter 12, where switching control between still image and moving image processing methods is performed, and the video signal of each pixel is Signal processing is performed according to the image or video.

In the above embodiment, the four pixels on the left, right, top, and bottom of the pixel of interest are used as surrounding pixels, but the present invention is not limited to this, and the positions and number of surrounding pixels can be arbitrarily set i!
It is something that can be enriched.

Further, in the embodiment, a delay element is used to select each determination signal of the pixel of interest and surrounding pixels, but this may be selected using a register or the like, for example.

(Effects of the Invention) As is clear from the above description, the motion signal detection circuit in the digital television receiver according to the present invention detects the level of each frame difference signal between the pixel of interest and its surrounding pixels, and Obtain a judgment signal of “dynamic” or “static”,
The final determination of whether the pixel of interest is dynamic or static is made by taking a majority vote of these determination signals.

Therefore, according to the present invention, low-level frame difference signals are not truncated as in the conventional example, and
Eliminates false motion signals caused by zinc and noise in video signals.
Since it can be removed with almost no effect on the original motion signal, the detection accuracy of the original motion signal can be improved.

Further, according to the present invention, since a motion signal is detected by majority decision of the determination signals of a plurality of pixels, there is no need for a complicated circuit configuration like the conventional motion signal detection circuit explained in FIG. The circuit scale can be reduced. Moreover, this also reduces power consumption, which is advantageous for digital television receivers, which are required to be downsized and power-saving as consumer equipment.

[Brief explanation of drawings]

FIG. 1 is a block diagram schematically showing the configuration of a motion signal detection circuit in a digital television receiver according to an embodiment of the present invention, FIG. 2 is an explanatory diagram of the operation of the embodiment, and FIG. FIG. 4 is a block diagram schematically showing the configuration of a frame difference signal detection circuit used in the example and the conventional example. FIG. 4 is a block diagram of a conventional motion signal detection circuit. 6... Frame difference signal detection circuit, 7... Dynamic/static judgment circuit, 8... Judgment signal extraction circuit, 9a, 9b...
- 1 scanning line delay element, 10a-10d...l sample delay element, 11...majority circuit, 12...3 sample low pass filter.

Claims (1)

[Claims] Frame difference signal detection means for detecting a frame difference signal for each pixel by calculating the difference in luminance signals between frames; and receiving the frame difference signal and comparing its level with a reference level. A motion/stillness determination means for determining whether each pixel is "moving" or "still"; and a determination signal is provided from the motion/still determination means, and the determination signals of the pixel of interest and a plurality of surrounding pixels are determined. A determination signal extracting means for extracting and outputting each determination signal, and a majority decision means for outputting a motion signal corresponding to "moving" or "static" of the pixel of interest based on a majority vote of each determination signal provided from the determination signal extracting means. A motion signal detection circuit for a digital television receiver, characterized in that: