JP3444170B2 - Field image interpolation method and field image interpolation device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a field image interpolation method and a field image interpolation device for interpolating another field image by one field image in an interlaced scanning type video signal.

[0002]

2. Description of the Related Art In an interlaced scanning type video signal, a multi-field image constitutes one frame image. Therefore, when a frame still image is obtained from the video signal, the frame still image is blurred due to a difference in image between adjacent fields. appear.

In particular, in a video signal of a fast-moving image, blurring becomes remarkable, and a high-quality frame still image cannot be obtained. Therefore, when the blurring is remarkable, the blurring is prevented by interpolating the other field image based on one field image of the plurality of field images forming one frame image.

FIG. 7 is a block diagram for explaining a field image interpolating apparatus using a conventional field image interpolating method for an interlace scanning type video signal. Here, 1 is a video signal input terminal to which a video signal of an interlaced scanning system is input, 2 is an A / D converter for converting the video signal input as an analog signal into a digital video signal, and 3 is the A / D A memory unit 4 for storing video data for one frame of the digital video signal output from the conversion unit based on a user's instruction is a field interpolating unit for producing a frame still image without blur by a method described later.

Reference numeral 5 is a timing at which a timing signal is sent to the A / D conversion unit 2, the memory unit 3 and the field interpolation unit 4 based on a synchronization signal in the video signal input from the video input terminal 1. The control unit 6 is a field interpolation instruction unit that outputs an instruction signal for interpolating a field image based on an instruction from the user to the field interpolation unit 4.

Here, a conventional field image interpolation method will be described. FIG. 8 shows a state of an image stored in the memory unit 3. In FIG. 8, for example, when the odd field image is interpolated based on the even field image, the pixel [m, n] to be interpolated is set to the adjacent pixel in the sub-scanning direction.
Replace with [m, n-1], or both adjacent pixels in the sub-scanning direction [m, n-
There is a method of obtaining from the average value of the brightness data of [1] and [m, n + 1].

Further, as another method, as shown in FIG. 9, adjacent pixels 3 in the sub-scanning direction centering on the pixel [m, n] to be interpolated are used.
Pixel by pixel, ie [m-1, n-1], [m, n-1], [m + 1, n-1], [m-1, n +
There is a method of obtaining interpolation data by deciding the interpolation axis by paying attention to a total of 6 pixels of 1], [m, n + 1], [m + 1, n + 1].

At this time, the brightness of each pixel is set to a, b, c, g,
In the case of displaying with h and i, in order to determine which direction of the L axis, the C axis, or the R axis to perform interpolation, the luminance difference between two pixels existing on each axis, that is, | c−g |, | B-
The values of h | and | a-i | are calculated, and the axis having the minimum value is determined as the interpolation axis. After that, the average value of the two pixels existing on the determined interpolation axis, that is, (c +
g) / 2 or (b + h) / 2, or (a + i) / 2
There is a method of obtaining interpolation data of [m, n] pixels by.

[0009]

However, the pixel [m, n] desired to be interpolated, which has been described with reference to FIG. 8, is replaced with the adjacent pixel [m, n-1] in the sub-scanning direction, or in the sub-scanning direction. Although it is possible to prevent blurring by the method of obtaining from the average value of the brightness data of both adjacent pixels [m, n-1] and [m, n + 1], there is an object in the diagonal direction on the pixel to be interpolated. In this case, the diagonal lines forming the contour of the object are not smooth lines but jagged lines (hereinafter referred to as jaggies) or unclear lines, and high-quality frame still images cannot be obtained.

Further, as described with reference to FIG. 9, the interpolation axis is determined by calculating the brightness difference between adjacent pixels, and the interpolation data of the pixel [m, n] is determined by the average value of the two pixels existing on the interpolation axis. Although it is possible to obtain a still image with less jaggies by using the method of obtaining, there is, for example, one thin diagonal line A as shown by a halftone dot in the brightness where the background is uniform, as shown in FIG. In that case, an incorrect interpolation axis may be selected, and an interpolation error occurs, leading to deterioration in image quality.

That is, in FIG. 10, the luminance c at the pixel [m + 1, n-1] and the luminance g at the pixel [m-1, n + 1] are interpolated for the pixel [m, n] to be interpolated. Since it is close to the power luminance data e, it is ideal to obtain the luminance data e by the average value of the luminance c and the luminance g. Therefore, the pixel [m + 1, n
-1] and [m-1, n + 1], that is, the L axis should be the interpolation axis, but two pixels on the L axis [m + 1, n-1] and [m-1 ,
2 pixels on the R axis [m-1, n-1] and [m +] rather than the brightness difference of [n + 1]
The brightness difference of [1, n + 1] may be small, in which case the R axis is selected as the interpolation axis.

FIG. 11 shows an example in which the image shown in FIG. 10 is interpolated using the R axis as an interpolation axis, and one thin diagonal line A in FIG. There is.

As a countermeasure against this, the pixel to be interpolated
There is also a method of accurately determining the interpolation axis by increasing the number of reference pixels, such as 5 pixels or 7 pixels, instead of referring to every 3 pixels in the sub-scanning direction centered on [m, n]. When the number of pixels is increased, there are disadvantages that the processing scale increases, the device cost increases, and the processing time increases.

[0014]

In order to solve the above problems, the field image interpolation method according to the present invention is based on the interlace scanning method in which a video signal of one frame is formed by the first and second fields. A field image interpolation method for obtaining an image information signal of the second field from an image information signal of the first field in a video signal, and obtaining interpolation data of a pixel [m, n] on the second field. On the first field, and on the second field
Pixel vertically adjacent to pixel [m, n] on the field
[m, n-1] and [m, n + 1] and a pixel [m, n-1] that is on the same field and is adjacent to the pixel [m, n-1] in the horizontal direction. m-1, n-1] and [m + 1, n-1] on the same field as the pixel [m, n + 1] and horizontally adjacent to the pixel [m, n + 1] Pixels [m-1, n + 1] and [m + 1, n + 1] are extracted as reference pixels, and based on the extracted reference pixel data, the pixels [m + 1, n-1] and [m-1, n + 1] in the first direction or the pixel [m-1, n-1] and [m + 1, n + 1] in the second direction. determining <br/> direction or the pixel [m, n-1] and [m, n + 1] of the third interpolation axis <br/> to any of direction towards connecting towards the Obtaining interpolated data of the pixel [m, n] based on the data of the reference pixel existing on the determined interpolation axis,
The determination of the interpolation axis is performed by the pixels [m + 1, n + 1], [m-1, n + 1],
The difference between the data of each reference pixel in the first region connecting [m + 1, n-1] and the pixels [m-1, n-1], [m + 1, n-1], [m −1, n + 1] and the difference between the data of the reference pixels in the second region, and the pixel
[m-1, n + 1], [m + 1, n + 1], [m-1, n-1] The difference between the data of each reference pixel in the third region connecting the pixel and the pixel [m + 1, n-1], [m-
1, n-1], [m + 1, n + 1], the difference between the data of the reference pixels in the fourth region and the pixels [m, n-1], [m, n + 1] comparing the difference between the data of the reference pixels in the fifth <br/> region connecting said first
Or, the difference in data of each reference pixel in the second area is the smallest
In the case of, the interpolation axis is determined to be the first direction,
The difference in the data of each reference pixel in the third or fourth region is
In the minimum case, the interpolation axis is determined to be the second direction,
The difference in the data of each reference pixel in the fifth region is the smallest
In the case of, the interpolation axis is determined to be the third direction.
It is characterized that, the data of the reference pixel is characterized by a luminance data, and the data of the reference pixel is to shall and characterized in that the color data.

Further, the field image interpolating apparatus according to the present invention is an interlace scanning type video signal in which a video signal of one frame is formed by the first and second fields, and the image information signal of the first field. A field image interpolating device for obtaining the image information signal of the second field from the memory means for storing the video signal of the one frame, and a pixel [m, n] on the second field.
When obtaining the interpolation data of, on the first field,
And pixels [m, n-1] and [m, n + 1] vertically adjacent to the pixel [m, n] on the second field, and the pixel [m, n-1]
Pixels [m-1, n-1] and [m + 1, n-1] that are on the same field as and that are adjacent to the pixel [m, n-1] in the horizontal direction,
Pixels [m-1, n + 1] and [m + 1, n + 1] on the same field as [m, n + 1] and horizontally adjacent to the pixel [m, n + 1] As reference pixels from the memory means, and based on the data of the extracted reference pixels, the pixels [m + 1, n-1] and [m-1,
n + 1] in the first direction or the pixels [m-1, n-1] and [m + 1, n +
1] in the second direction, or the pixels [m, n-1] and [m, n +]
1] interpolating axis determining means for determining an interpolation axis in any of the third directions, and interpolation of the pixel [m, n] based on the data of the reference pixel existing on the determined interpolation axis. Interpolation interpolation means for obtaining data, and the interpolation axis determination means connects the pixels [m + 1, n + 1], [m-1, n + 1], [m + 1, n-1] The difference between the data of each reference pixel in the area and the pixels [m-1, n-1], [m +
1, n-1], [m-1, n + 1] in the area connecting the reference pixels and the pixels [m-1, n + 1], [m + 1, n + 1] ], [M-1, n-1], the difference between the data of each reference pixel in the area connecting the pixel [m +
1, n-1], [m-1, n-1], [m + 1, n + 1] data difference of each reference pixel in the region connecting the pixel [m, n-1], The difference between the data of each reference pixel in the area connecting [m, n + 1] is compared , and the first or
Is the minimum difference between the data of the reference pixels in the second area.
In this case, the interpolation axis is determined to be the first direction,
The difference in the data of each reference pixel in the 3rd or 4th area is the largest.
If it is small, the interpolation axis is determined to be the second direction, and
Difference data of each referenced pixel in the serial fifth region is smallest
In this case, the interpolation axis is determined to be the third direction.
The was that characterized by the data of the reference pixels, and characterized in that the luminance data, and the data of the reference pixel is to shall and characterized in that the color data.

[0016]

1 is a block diagram for explaining a field image interpolation apparatus using a field image interpolation method according to an embodiment of the present invention. In FIG. 1, a video signal input terminal 1, an A / D conversion unit 2, a memory unit 3, a timing control unit 5, and a field interpolation instruction unit 6 have the same configuration as that of a conventional field image interpolation device using a field image interpolation method. Therefore, detailed description is omitted.

Here, 10 is a field interpolation unit,
The interpolation axis determination unit 10a and the interpolation calculation unit 10b are included. The interpolation axis determination unit 10a determines the interpolation axis at the timing of the timing signal from the timing control unit 5 based on the video data for one frame input from the memory unit 3.

Further, the interpolation axis calculation unit 10b has information of the interpolation axis determined by the interpolation axis determination unit 10a, one frame of video data from the memory unit 3, and a timing signal from the timing control 5. Also, an instruction signal from the field interpolation instruction unit 6 is input and a frame still image is output.

Next, the operation of the field image interpolation apparatus using the field image interpolation method according to the embodiment of the present invention will be described. The video signal input as an analog signal to the video signal input terminal 1 is converted into a digital video signal by the A / D converter 2. At this time, when the user does not select the mode for capturing the still image, the converted digital video signal is converted into an analog video signal by the D / A converter (not shown) and output from the output terminal (not shown).

On the other hand, when the user selects the mode for capturing a still image, the video data for one frame of the digital video signal output from the A / D conversion unit 2 is loaded into the memory unit 3 and the loading is performed. The video data for one frame is output to the printing means and the image display means (not shown) via the field interpolation unit 10.

At this time, the user can confirm the frame still image in which the video data for one frame output from the memory unit 3 is output as it is, by the image display means (not shown), and a good still image is displayed. If not, the field interpolation instructing section 6 gives an instruction to cause the field interpolating section 10 to interpolate the video signal.

When performing the interpolation processing, the reference pixels necessary for the interpolation are stored in the memory unit 3 and the interpolation axis determining unit 10a.
Are sequentially taken in to determine the interpolation axis, and the interpolation calculation unit 10b calculates the interpolation data and outputs it. In addition,
Timing of signal input / output and interpolation processing in the above configuration is controlled by a timing signal from the timing control unit 5.

As described above, in the fold image interpolating apparatus using the field image interpolating method according to the embodiment of the present invention, the interpolation axis determining unit 10a determines the interpolation axis when the data of the pixel to be interpolated is obtained. Here, first, an outline of the interpolation method will be described.

As described above, FIG. 10 shows the case where one thin diagonal line exists on the pixel [m, n] to be interpolated. For example, pixels [m-1, n-1], [m, n-1], [m + 1, n-1], [m
−1, n + 1], [m, n + 1], [m + 1, n + 1] represents the brightness of a,
In the case of b, c, g, h, i, the interpolation axis is pixel [m +
The L axis connecting 1, n-1] and [m-1, n + 1] is selected, and the interpolation data should be obtained by the average value of the luminances c and g.

However, when there is one thin diagonal line A on the image [m, n] to be interpolated, the luminances a, b, c, g,
It is very difficult to set the interpolation axis as the L axis from h and i. On the other hand, when the R axis is used as the interpolation axis as described above, the brightness e at the pixel [m, n] to be interpolated becomes the average value of the brightness a or i, and the thin diagonal line A appears like a wavy line in some places. However, if the C axis is the interpolation axis,
It is an average value of the brightnesses b and h, and although accurate interpolation data as when interpolation is performed with the L axis as an interpolation axis cannot be obtained, a still image that is good to some extent can be obtained.

Therefore, for one thin diagonal line existing on the pixel [m, n] to be interpolated, the C axis is used as the interpolation axis to obtain a certain amount of good interpolated data and the pixel to be interpolated [ A good interpolation axis can be set even when an object exists in the diagonal direction on [m, n].

As a concrete method, as shown in FIG.
In order to obtain the interpolation data e in the pixel [m, n] to be interpolated, first consider the triangular area shown in FIGS. 2A to 2D and the rectangular area shown in FIG.

In FIG. 2A, the reference pixels [m + 1, n-1], [m-1, n + 1], [m, n + 1] and [m + 1] existing in the triangular area. , n + 1] luminance data c, g, h and i are compared to calculate the data of the largest luminance difference. Similarly, in the triangular areas shown in FIGS. 2B to 2D, the luminance data of each reference pixel is compared, and the data of the largest luminance difference is calculated. Finally Figure 2
Pixels [m, n-1] existing in the rectangular area shown in (e) and
The brightness difference between the brightness data b and h of [m, n + 1] is calculated.

Then, the brightness differences calculated from the triangular and rectangular areas are compared, and if the brightness difference data calculated from the triangular area shown in (a) or (b) is the minimum value, the L-axis is calculated. , (C) or (d), the R axis when the brightness difference data calculated from the triangular area is the minimum value, (e)
When the brightness difference data calculated from the rectangular area shown in (1) is the minimum value, the C axis is used as the interpolation axis.

At this time, the luminances c and g in FIG.
Is the same as the luminance data for the same thin diagonal line A, and therefore has almost the same value, whereas the luminance i is not the luminance data for the thin diagonal line A, and therefore has a larger value than the luminances c and g. As the brightness difference data of the triangular area shown in FIG. 2A, a relatively large value is obtained.

Similarly, when considering FIGS. 2B to 2D, at least a pixel showing the luminance data of the thin diagonal line A and a pixel not showing the luminance data of the thin diagonal line A are at least in the triangular region. Since there is one each, a relatively large value can be obtained as the brightness difference data of each triangular area.

On the other hand, since the brightness data b and h in FIG. 2 (e) have almost the same value, the brightness difference data in the rectangular area has a small value, and comparing these values results in the C-axis. Will be chosen as the interpolation axis.

Further, as shown in FIG. 3, assuming that an object exists in an oblique direction as indicated by halftone dots with the L axis as a contour, the luminances c, g, h in FIG. Since i is the brightness data regarding the same object, i has almost the same value, and the value of the brightness data to be output is small, whereas i is in the regions shown in FIGS. 3B to 3E. Since there is at least one pixel indicating the luminance data of the object in the diagonal direction and at least one pixel not indicating the luminance data of the object in the diagonal direction, the value of the luminance data output from each region is large to some extent. As a result, the L axis is selected as the interpolation axis.

Although not shown here, the R axis is similarly selected as the interpolation axis when an object exists in an oblique direction with the R axis as the contour. After this,
As for the interpolation data e of the pixel [m, n], the average value of the brightness data of the two pixel data on the determined interpolation axis may be obtained, as described above.

The flow of the field image interpolation method according to the embodiment of the present invention will be described below. FIG. 4 is a flow chart showing a field image interpolation method according to an embodiment of the present invention.

First, adjacent 6 pixels [m-1, n-1], [m, n-1], [m + 1, n-1 in the sub-scanning direction centering on the pixel [m, n] to be interpolated. ],
The brightness a of each of [m-1, n + 1], [m, n + 1], and [m + 1, n + 1],
For b, c, g, h and i, c and g, c and h, c
And i are compared in brightness, and the data having the largest brightness difference is output as L1 (step 101). Next, the brightness differences of g and a, g and b, g and c are compared, and the data with the largest brightness difference is output as L2 (step 10).
2).

Then, the data of L1 is compared with the data of L2, and the smaller data is output as L (step 103). Here, the data L indicates the degree of existence of an object in an oblique direction with the L axis as a contour. That is, assuming that there is not much difference in brightness on the same object, the value of L should be small if the object exists in an oblique direction with the L axis as the contour.

Similarly, the brightness differences of a and g, a and h, a and i are compared, and the data with the largest brightness difference is output as R1 (step 104), and i and a, i and b, i And c are compared, and the data having the largest luminance difference is output as R2 (step 105). Then, the data of R1 and the data of R2 are compared, and the smaller data is output as R (step 1
06), using the R axis as a contour, the degree of presence of an object in an oblique direction is checked.

Then, data L and b indicating the degree of existence of an object in the oblique direction with the L axis as the contour and data R and b indicating the degree of existence of an object in the oblique direction with the R axis as the contour. And the luminance difference C of h (step 108), and the interpolation axis is determined by the data having the smallest value (step 109).

Here, when an object exists in an oblique direction with the L axis or the R axis as the contour, the value of the data L or the value of R becomes the smallest, so the interpolation axis is set to the L axis or the R axis. As shown in FIG. 10, when there is one thin diagonal line on the interpolated pixel, the brightness difference between b and h is small and the data C
Since the value of becomes smaller and the value of data L and the value of R become larger, the interpolation axis is set to the C axis.

Next, another field image interpolation method according to the embodiment of the present invention will be described. FIG. 5 is a flowchart showing another field image interpolation method according to the embodiment of the present invention.

First, six adjacent pixels [m-1, n-1], [m, n-1], [m + 1, n-1 in the sub-scanning direction centering on the pixel [m, n] to be interpolated. ],
The brightness of each of [m-1, n + 1], [m, n + 1], and [m + 1, n + 1] is a,
If b, c, g, h, i, then c and g, c and h,
The sum of the brightness differences of c and i is output as SL1 (step 201). Next, the sum of the brightness differences of g and a, g and b, g and c is output as SL2 (step 20).
2).

Then, the data of SL1 and the data of SL2 are compared (step 203). If the data of SL1 is small, the brightness differences of c and g, c and h, and c and i are compared to obtain the highest brightness. Data with a large difference is output as L (step 204), and if the data in SL2 is small,
The brightness differences of g and a, g and b, g and c are compared, and the data with the largest brightness difference is output as L (step 2
05). Here, the data L indicates the degree of existence of an object in an oblique direction with the L axis as a contour.

Similarly, the sum of the brightness differences of a and g, a and h, a and i is output as SR1 (step 206), and the sum of the brightness differences of i and a, i and b, i and c is calculated. It is output as SR2 (step 207). Then, the data of SR1 and the data of SR2 are compared (step 208). If the data of SR1 is small, the brightness differences of a and g, a and h, a and i are compared, and the brightness difference is the largest. Output the data as R (step 20
9) If the SR2 data is small, the brightness differences of i and a, i and b, i and c are compared, and the data with the largest brightness difference is output as R (step 210).

Here, the data R indicates the degree of existence of an object in an oblique direction with the R axis as a contour. Then, data L indicating the degree of existence of an object in the oblique direction with the L axis as the contour and data R indicating the degree of existence of an object in the oblique direction with the R axis as the contour, and b and h. The brightness difference C is compared (step 212), and the interpolation axis is determined by the data having the smallest value (step 213).

Here, when an object exists in an oblique direction with the L axis or the R axis as the contour, the value of the data L or the value of R becomes the smallest, so the interpolation axis is set to the L axis or the R axis. As shown in FIG. 10, when one thin diagonal line exists for the interpolated pixel, the brightness difference between b and h is small, the value of the data C is small, and the values of the data L and R are small. Since it becomes larger, the interpolation axis is set to the C axis.

By determining the interpolation axis in this way, even if there is one thin diagonal line on the pixel [m, n] to be interpolated, the L axis or the R axis is used as the contour. Even if there is an object in an oblique direction, the possibility of selecting an incorrect interpolation axis that causes an interpolation error and is prominent in image quality deterioration is reduced.

As shown in FIG. 6, the video data stored in the memory unit 3 is processed by the field interpolation unit 1 for interpolation processing.
It is also possible to configure so that the video data is output to 0 and the video data is returned to the memory unit 3 after the interpolation processing is completed in the field interpolation unit 10. In the case of such a configuration, the data after the interpolation processing is output from the memory unit 3 to the printing unit or the image display unit (not shown), and the interpolation processing speed in the field interpolation unit 10 is slow. Even in this case, the data can be output regardless of the processing speed.

Further, in the above description, an example in which the interpolation data is calculated based on the luminance signal when obtaining the interpolation data is shown, but it goes without saying that the same applies when a color signal or a gradation signal is used.

[0050]

As described above, according to the present invention, even when one thin diagonal line exists on the pixel [m, n] to be interpolated, the L axis or the R axis is used as a contour in the diagonal direction. Even if there is an object in the image, it is unlikely to select an incorrect interpolation axis that causes an interpolation error and causes image quality deterioration, so that a good frame still image can be obtained.

Further, since the interpolation data can be obtained with a small number of reference pixels, the processing scale can be reduced, which is advantageous in terms of apparatus cost and processing time.

[Brief description of drawings]

FIG. 1 is a block diagram for explaining a field image interpolation device using a field image interpolation method according to an embodiment of the present invention.

FIG. 2 is a diagram for explaining an outline of a field image interpolation method according to an embodiment of the present invention.

FIG. 3 is a diagram showing a state in which an object exists in an oblique direction with the L axis as a contour.

FIG. 4 is a flowchart showing a field image interpolation method according to an embodiment of the present invention.

FIG. 5 is a flowchart showing another field image interpolation method according to an embodiment of the present invention.

FIG. 6 is a diagram for explaining another field image interpolation device using the field image interpolation method according to the embodiment of the present invention.

FIG. 7 is a block diagram for explaining a field image interpolation apparatus using a conventional field image interpolation method.

FIG. 8 is a diagram showing a state of an image stored in a memory unit.

FIG. 9 is an explanatory diagram for explaining a conventional field image interpolation method.

FIG. 10 is a diagram showing a state in which one thin diagonal line exists on a pixel to be interpolated.

FIG. 11 is a diagram showing a state in which an interpolation error has occurred.

[Explanation of symbols]

1 ... Video signal input terminal 2 ... A / D converter 3 ... Memory section 4, 10 ... Field interpolator 5 ... Timing control section 6 ... Field interpolation instruction section 10a ... Interpolation axis determination unit 10b ... Interpolation calculation unit

Claims (6)

(57) [Claims] 1. An interlace scanning type video signal in which a video signal of one frame is formed by the first and second fields, and from the image information signal of the first field to the image information signal of the second field. Is a field image interpolation method for obtaining the pixel [m, n] on the second field and the pixel [m, n] on the second field when obtaining the interpolation data of the pixel [m, n] on the second field. [n,] vertically adjacent pixels [m, n-1] and
[m, n + 1] on the same field as the pixel [m, n-1] and the pixel
Pixels [m-1, n-1] and [m + 1, n] that are horizontally adjacent to [m, n-1]
-1] and the pixel [m, n + 1] on the same field and the pixel
Pixels [m-1, n + 1] and [m + 1, n] that are horizontally adjacent to [m, n + 1]
+1] as a reference pixel, and based on the data of the extracted reference pixel, the pixel [m +
1, n-1] and [m-1, n + 1] in the first direction or the pixel [m-1,
the second direction connecting [n-1] and [m + 1, n + 1], or the pixel
determining an interpolation axis in any of the third directions connecting [m, n-1] and [m, n + 1], and based on the data of the reference pixel existing on the determined interpolation axis. To obtain interpolation data of the pixel [m, n], the interpolation axis is determined by the pixel [m + 1, n + 1], [m-1, n + 1],
The difference between the data of each reference pixel in the first region connecting [m + 1, n-1] and the pixels [m-1, n-1], [m + 1, n-1], [m −1, n + 1] and the difference between the data of the reference pixels in the second region, and the pixel
[m-1, n + 1], [m + 1, n + 1], [m-1, n-1] The difference between the data of each reference pixel in the third region connecting the pixel and the pixel [m + 1, n-1], [m-
1, n-1], [m + 1, n + 1], the difference between the data of the reference pixels in the fourth region and the pixels [m, n-1], [m, n + 1] comparing the difference between the data of the reference pixels in the fifth <br/> region connecting said first
Or, the difference in data of each reference pixel in the second area is the smallest
In the case of, the interpolation axis is determined to be the first direction,
The difference in the data of each reference pixel in the third or fourth region is
In the minimum case, the interpolation axis is determined to be the second direction,
The difference in the data of each reference pixel in the fifth region is the smallest
In the case of, the interpolation axis is determined to be the third direction.
Field image interpolation method being characterized in that the. 2. The field image interpolation method according to claim 1, wherein the reference pixel data is luminance data. 3. The field image interpolation method according to claim 1, wherein the reference pixel data is color tone data. 4. An interlace scanning type video signal in which a video signal of one frame is formed by the first and second fields, and from the image information signal of the first field to the image information signal of the second field. And a memory means for storing the video signal of the one frame, and interpolating data of the pixel [m, n] on the second field, in the first field. And a pixel [m, n-1] vertically adjacent to the pixel [m, n] on the second field and
[m, n + 1] on the same field as the pixel [m, n-1] and the pixel
Pixels [m-1, n-1] and [m + 1, n] that are horizontally adjacent to [m, n-1]
-1] and the pixel [m, n + 1] on the same field and the pixel
Pixels [m-1, n + 1] and [m + 1, n] that are horizontally adjacent to [m, n + 1]
+1] as a reference pixel from the memory means, and based on the data of the extracted reference pixel, the pixel [m +
1, n-1] and [m-1, n + 1] in the first direction or the pixel [m-1,
the second direction connecting [n-1] and [m + 1, n + 1], or the pixel
interpolation axis determining means for determining an interpolation axis in any of the third directions connecting [m, n-1] and [m, n + 1], and the reference pixel existing on the determined interpolation axis. An interpolation calculation unit that obtains interpolation data of the pixel [m, n] based on data, and the interpolation axis determination unit includes the pixels [m + 1, n + 1], [m-1, n +
1] and [m + 1, n-1] in the area connecting the reference pixels and the pixels [m-1, n-1], [m + 1, n-1], [m −1, n + 1] and the difference between the data of each reference pixel in the area connecting
1], [m + 1, n + 1], [m-1, n-1], the difference between the data of the reference pixels in the area connecting the pixels [m + 1, n-1], [m -1, n-1], [m + 1, n
+1] and the difference between the data of the reference pixels in the region connecting the pixel [m, n-1], compares the difference between the data of the reference pixels in the region connecting the [m, n + 1] , Each of the references in the first or second areas
When the difference in the data of the illuminated pixels is the minimum, the interpolation axis is set to
In the first or third direction, in the third or fourth area
If the difference in the data of each reference pixel is the minimum, the interpolation axis is set to
Determined as the second direction, each reference in the fifth area
When the difference in the data of the illuminated pixels is the minimum, the interpolation axis is set to
A field image interpolating device, characterized in that the third direction is determined . 5. The field image interpolating device according to claim 4 , wherein the data of the reference pixel is luminance data. 6. The field image interpolating apparatus according to claim 4 , wherein the data of the reference pixel is color tone data.