JP3435732B2 - Video signal filtering method and video special effect device using the same - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video special effect device and a video signal for use in creating an image which has been subjected to processing such as enlargement, reduction and rotation based on an original image by utilizing image processing technology. The present invention relates to a filter processing method.

[0002]

2. Description of the Related Art A conventional example of a video special effect device is T
Some are listed in V-Tech IPA87-3. That is, by using the digital image processing technique, the data of each pixel of the digitized input image is subjected to an operation corresponding to a process of performing shape conversion such as enlargement, reduction, rotation, or three-dimensional mapping. , Special effects video (Iida, Okada, Nishimura et al .: Real-time 3D special effects device).

As such a conventional special effect device,
As shown in FIG. 4, an input terminal 1 to which a digitized video signal is input, a filter processing circuit 2 that filters the video signal of each pixel, and an output video signal for the input video signal of each pixel. Of the horizontal and vertical directions to generate a reduction ratio and provides it to the filter processing circuit 2, a frame memory 4 for storing the video signals output from the filter processing circuit 2 in raster scan order, and a frame memory. An address generation circuit 5 for generating an address on the memory of a video data signal to be read according to conversion such as reduction, rotation, enlargement, etc. for the video signal stored in 4.
And an interpolation processing circuit 6 for randomly reading a video data signal from the frame memory 4 according to an address provided by the address generation circuit 5, and an output terminal 7 for outputting the video signal subjected to the interpolation processing.

In the special effect device thus constructed, the video signal for each pixel input from the input terminal 1 is input to the filter processing circuit 2. When reducing the image, aliasing distortion (alias
In order to prevent the ng), a vertical frequency band is generated by the filter processing circuit 2 in accordance with the horizontal and vertical reduction ratio of each pixel of the output video signal with respect to the input video signal provided from the reduction ratio generation circuit 3. To limit. The horizontally and vertically filtered video signals are temporarily stored in the frame memory 4 and then, in the interpolation processing circuit 6, a shape such as reduction, rotation, enlargement or three-dimensional mapping provided by the address generation circuit 5. The video data signal corresponding to the address corresponding to the conversion process is read out and output from the output terminal 7.

Since the address provided by the address generating circuit 5 usually has a small number of parts, the input pixels corresponding to the output pixels do not correspond one-to-one. For this reason, the interpolation processing circuit 6 uses the decimal part of the address to generate output pixels from pixels near the input video through interpolation processing.

Further, as an interpolation processing circuit of the special effect device, there has been proposed a method of changing the interpolation point by using the motion information of the input video signal in order to improve the resolution in the vertical direction. (Kaisho 63-121364).

As shown in FIG. 5, such a special effect device has an input terminal 1 to which a digitized video signal is input.
And a frame memory 4 for storing two consecutive fields in raster scan order, and a video data signal to be read according to conversion such as reduction, rotation, enlargement, etc. for the video signal stored in the frame memory 4 on the memory. An address generation circuit 5 for generating an address, a motion detection circuit 8 for detecting motion information of an input video signal, and a frame memory randomly according to an address provided by the address generation circuit 5 and motion information provided by the motion detection circuit 8. An interpolation processing circuit 6 for reading a video data signal from the output terminal 4 and an output terminal 7 for outputting the video signal subjected to the interpolation processing.

In the special effect device thus constructed, the video signal for each pixel input from the input terminal 1 is stored in the frame memory 4 for two consecutive fields. The motion detection circuit 8 outputs motion information indicating whether there is motion between the frames of the input video signal. In the interpolation processing circuit 6, a signal of image data corresponding to the address and the motion information provided by the address generation circuit 5 corresponding to shape conversion processing such as reduction, rotation, enlargement or three-dimensional mapping is read from the frame memory 4. And output from the output terminal 7.

Since the address provided by the address generation circuit 5 usually has a small number of parts, the input pixels corresponding to the output pixels do not correspond one-to-one. For this reason, the interpolation processing circuit 6 uses the decimal part of the address to generate output pixels from pixels near the input video through interpolation processing. At this time, when it is determined that the motion information is “motion” in the pixels in the vicinity indicated by the address provided by the address generation circuit 5, the interpolation processing circuit 6 uses the adjacent pixels of the input image in the same field. To perform interpolation processing. Further, when it is determined that the motion information image is “still” in the pixels near the address indicated by the address provided by the address generation circuit 5,
The interpolation processing circuit 6 performs the interpolation processing on the continuous input video of two fields by using the neighboring pixels indicated by the address.

As described above, when the motion information is determined to be "still", the output image is obtained from the continuous input image of two fields, as compared with the case where the output image is obtained from the input image of one field. Good vertical resolution is obtained.

[0011]

However, in the above-mentioned conventional special effect device, the filter processing at the time of the reduction processing is reduced to one.
Since this is performed only on the input image in the field, there are problems that aliasing distortion occurs in the stationary portion and vertical resolution deteriorates.

This problem will be described with reference to FIG. 6 by taking as an example the case where the vertical size is reduced to ½.

The horizontal axis of FIG. 6 is the vertical frequency, and the vertical axis is the gain. Assuming that the number of vertical scanning lines is 525 of NTSC, the vertical frequency that can be expressed is 525/2 [cp].
H] (cycle / Height (screen height)).

In the case of a still image, a band may exist as shown in FIG. 6 (a). At this time, like the conventional special effect device,
If a filtering process that limits the vertical band to half is performed on the input image of the field in order not to generate aliasing distortion, a high frequency band passes like a shaded portion in FIG. 6B, When shrinking, foldback distortion may occur,
This causes deterioration of the vertical resolution.

The present invention solves such a conventional problem, and eliminates aliasing distortion in a stationary portion when reduction is performed, and improves vertical resolution, thereby improving image quality by filtering a video signal. An object is to provide a processing method and a special effect device using the same.

[0016]

To achieve the above object, according to the Invention The filter processing method of the video signal of the present invention is a filter characteristic corresponding to the vertical and horizontal reduction ratio of the output image with respect to the input video signal filter a filtering method for performing processing, the target pixel in the input video signal to obtain a motion information by the motion or still, the motion information
When is “Still”, the image of the current field and the previous field is
The vertical direction that limits the band of the input video signal with respect to the element
If the motion information is “motion” after filtering,
Band of the input video signal only for pixels in the current field
The vertical filtering process for limiting

Further, the video special effect apparatus of the present invention comprises a motion detection circuit for detecting motion information due to the motion or stillness of the pixel of interest in the input video signal, and a filter processing circuit for filtering the input video signal. , The video data is read from a frame memory that records two consecutive fields at a predetermined address with respect to the video signal output from the filter processing circuit, an interpolation process for performing a video conversion process is performed, and the motion information is converted into the motion information. Is "stationary"
In the case of, the interpolation processing is performed using the video data of the current field and the previous field, and when the motion information is “motion”,
An interpolation processing circuit for performing an interpolation process using the video data in the current field; a means for providing the filtering circuit with vertical and horizontal reduction ratios of the output video signal with respect to the input video signal; A means for providing information to the filter processing circuit is provided, and the filter processing circuit performs filter processing of filter characteristics according to the scale factor.
If performed, and if the motion information is "static", to filter the vertical limiting the band of the input video signal to the pixels of the current field and the previous field, the motion information is "motion", The vertical filtering process for limiting the band of the input video signal is performed only on the pixels of the current field.

[0018]

Therefore, according to the method of the present invention, since the filtering process is performed on the points set according to the motion information of the input video signal, the aliasing distortion in the stationary portion when the reduction is performed is eliminated, The vertical resolution can be improved.

Further, in the video special effect apparatus having the above-mentioned structure, the filtering process of the filter characteristic according to the vertical and horizontal reduction ratios of the input video signal is performed in the point corresponding to the motion information detected by the motion detecting means. As a result, the aliasing distortion in the stationary portion when the image is reduced is eliminated, the vertical resolution is improved, a sharp image can be output, and the image quality can be improved.

[0020]

EXAMPLES Examples of the present invention will be described below with reference to FIGS.

FIG. 1 is a block diagram showing the structure of a video special effect device using the filter processing method according to the present invention. The input terminal 1, similar to the case shown in FIG. 4 and FIG.
The filter processing circuit 2, the reduction rate generation circuit 3, the frame memory 4, the address generation circuit 5, the interpolation processing circuit 6, the output terminal 7 and the motion detection circuit 8 are provided, and the conventional example shown in FIGS. The difference is that the motion information of the input video signal for each pixel, which is obtained from the motion detection circuit 8, is supplied to the filter processing circuit 2, and that the motion processing is used to set the point to be filtered by the filter processing circuit 2. It is in the place where it was done.

Next, the operation of this embodiment configured as described above will be described. The video signal (digital signal) for each pixel input from the input terminal 1 is filtered by the filter processing circuit 2. At this time, the filter characteristic in the filter processing circuit 2 is set by the horizontal and vertical reduction ratios of the output video signal with respect to the input video signal provided by the reduction ratio generation circuit 3. Then, the points to be filtered by the filtering circuit 2 are set by the motion information provided by the motion detection circuit 8. The filtered horizontal and vertical video signals are temporarily stored in the frame memory 4 and then the address according to the processing such as reduction, rotation, enlargement provided by the address generation circuit 5 and the motion detection circuit 8 are performed. Video data corresponding to the provided motion information is read from the frame memory 4 in the interpolation processing circuit 6, subjected to interpolation processing, and then output from the output terminal 7.

Next, in the vertical filtering process, when the input image is reduced in the vertical direction at the same reduction ratio of the entire screen, the point to be filtered is determined according to the motion information provided by the motion detection circuit 8. How to set is described below.

A one-dimensional FIR digital filter is used for the filter processing circuit 2 of this embodiment. In this case, the motion information provided by the motion detection circuit 8 is “still” or
The filter processing points corresponding to the motion information when "motion" is shown in FIG.

FIG. 2 shows the filtering points in the vertical direction when the tap length of the filter is 5 taps.
When the motion information of the pixel of interest is “still”, the filtering process is performed on the pixels of the current field and the previous field as shown in FIG. If the information of the pixel of interest is “motion”,
As in (b), the filtering process is performed only on the pixels in the current field.

Such processing can be realized by a commonly used FIR filter by simply switching the filter processing points according to the motion information for the target pixel of the filter processing.

Further, in the present embodiment, when the still image is subjected to the reduction processing of 1/2 in the vertical direction, the frequency characteristic in the filter processing will be described with reference to FIG.

FIG. 3 (a) is the same as FIG. 6 (a), and the expressible frequency is 525/2 [cpH].
At this time, when the pixel processing as shown in FIG. 2A is subjected to the filter processing for limiting the vertical band to half, the band limitation as shown in FIG. 3B is performed. In this case, the high frequency band that causes the aliasing distortion as shown in FIG. 6B is band-limited, and aliasing distortion does not occur even when the size is reduced.

As a result, in the video special effect apparatus of this embodiment, the filtering process of the filter characteristic corresponding to the vertical reduction ratio and the horizontal reduction ratio corresponding to the input video is set according to the motion information. By doing this for the points, there is no aliasing distortion in the static part when reducing,
In addition to improving the vertical resolution, it is possible to output sharp images and improve the image quality.

In the filter processing in the horizontal direction, it is not necessary to change the filter processing point according to the motion information.

Further, here, the filter processing points are changed according to the motion information of only the target pixel to be filtered, but the filter processing points are changed by using a lot of motion information for the tap length to be filtered. May be.

In the filtering process, the vertical filtering process may be performed after the horizontal filtering process, or the horizontal filtering process may be performed after the vertical filtering process.

[0033]

As is apparent from the above embodiment, the present invention sets the filter processing of the filter characteristic according to the vertical reduction ratio and the horizontal reduction ratio corresponding to the input video according to the motion information. By doing so, the aliasing distortion in the stationary part when the image is reduced is eliminated, the vertical resolution can be improved, and a sharp image can be output, and the image quality of the image special effect device is improved. It has the effect of being able to.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of a video special effect device to which the method of the present invention is applied.

FIG. 2 is a diagram showing vertical filtering points in the embodiment of the present invention.

FIG. 3 is a characteristic diagram showing vertical filter characteristics according to an embodiment of the present invention.

FIG. 4 is a block diagram showing a configuration of a conventional video special effect device.

FIG. 5 is a block diagram showing the configuration of a conventional video special effect device.

FIG. 6 is a characteristic diagram showing vertical filter characteristics in a conventional image special effect device.

[Explanation of symbols]

1 input terminal 2 Filter processing circuit 3 Reduction rate generation circuit 4 frame memory 5 Address generation circuit 6 Interpolation processing circuit 7 output terminals 8 Motion detection circuit

─────────────────────────────────────────────────── ─── Continuation of the front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) H04N 5/262

Claims (2)

(57) [Claims] 1. A filtering method for performing filter processing of the filter characteristics corresponding to vertical and horizontal reduction ratio of the output image with respect to input video signal, the target pixel in the input video signal, a motion or still If the motion information is “still”, the current feel is obtained.
Band of the input video signal for the pixels of
Vertical motion filtering that limits the area
If the information is "motion", only the pixels in the current field
Vertical filter for limiting the band of the input video signal
A method of filtering a video signal, characterized by performing processing. 2. A motion detection circuit for detecting motion information due to a motion or stillness of a pixel of interest in an input video signal, a filter processing circuit for filtering the input video signal, and an output video signal of the filter processing circuit. In contrast, when the video data is read from the frame memory that records two consecutive fields at a predetermined address, an interpolation process for performing a video conversion process is performed, and when the motion information is “still”, the current field is read. And an interpolation processing circuit that performs interpolation processing using the video data of the previous field and the motion information is “motion”, and an interpolation processing circuit that performs the interpolation processing using the video data of the current field and the input video signal. Means for providing the filtering circuit with vertical and horizontal reduction ratios of the output video signal, and the motion information with the filtering process. And means for providing sense circuit is provided, wherein the filtering circuit, a filter especially in accordance with the scaling factor
Filtering, and the motion information is "still"
In the case of, the vertical filtering process for limiting the band of the input video signal is performed on the pixels of the current field and the previous field. A video special effect device characterized by performing vertical filtering processing for limiting a signal band.