JP2006060375A - Sequential-scanning converter - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a program for sequential-scanning conversion capable of conducting an accurate interpolation in the interpolation in a field and capable of improving the picture quality of a picture after the interpolation and a device for the program. <P>SOLUTION: A sequential-scanning converter has an interpolation-region setter 3 determining an interpolation-element region using an interpolation element as a reference in a frame memory 1 storing an interlaced video signal and an interpolation-direction dependency-quantity computer 4 obtaining the quantity of the dependency of the direction Di to the interpolation element from the coefficient ki of the directions (D1, D2,...) preset to a memory 8 to the interpolation element and the memory 6. The sequential-scanning converter further has an interpolation-direction determining section 5 collecting the quantities of the dependency (D1, D2,...) obtained by the interpolation-direction dependency-quantity computer 4 and determining the optimum interpolation direction from the result of the collection, and an interpolation processor 7 using picture elements based on the interpolation direction determined by the interpolation-direction determining section 5 and the average of the picture elements as the values of the interpolation element of the frame memory 1. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a program and an apparatus for interpolating interpolation elements when converting a moving image signal or the like of interlaced scanning (interlaced scanning) into a moving image signal of progressive scanning (progressive scanning) (IP conversion). .

  In general, a moving image signal has two types of scanning structures, interlaced scanning and progressive scanning. In interlaced scanning, the number of scanning lines is reduced to half of the number of scanning lines in sequential scanning.

  In television broadcasting, interlaced scanning is generally used, but in digital broadcasting, sequential scanning is also used. A display for a personal computer (PC) is sequentially scanned to display fine characters.

  On the other hand, since a display having a matrix type display driver such as a plasma display or a liquid crystal display is difficult to perform interlaced scanning display, sequential scanning display is performed.

  Also, CRT television receivers with HDTV (high-definition) display systems, when trying to display current broadcast images with a small number of scanning lines, will cause the deflection frequency to be too low with interlace scanning. In many cases, images are displayed by sequential scanning with a doubled deflection frequency.

  Conventionally, there are the following interpolation methods in the field for IP conversion.

A) Interpolation is performed with the average value of the upper and lower line pixels as shown in FIG.

B) Interpolation is performed using the correlation in the oblique direction in the field as shown in FIG.
Japanese Patent Laid-Open No. 4-355581

  However, the background art as described above has the following disadvantages.

(1) Method (a) As shown in FIG. 16, since interpolation is based only on the upper and lower pixel average values, accurate interpolation is performed in an image portion where the vertical frequency is high, that is, an image portion where the vertical direction changes rapidly. Can not be performed, the diagonal line becomes jagged.

(2) Method (a) As shown by Db and Dd in FIG. 17A, the correlation in the oblique direction is detected at a position two pixels away in the horizontal direction. According to this method, in the case of an oblique line having a width of two or more horizontal pixels as shown in FIG. 5B, the correlation in the same direction as the oblique line is reduced as indicated by an arrow in the figure. Therefore, correct interpolation is possible.

  However, in the case of an oblique line having a width of less than two horizontal pixels as shown in FIG. 5C, the correlation between the oblique line and the opposite direction becomes small as indicated by an arrow in the figure. For this reason, erroneous interpolation will be performed.

  Therefore, the present invention focuses on the above points, and a program for sequential scan conversion capable of performing accurate interpolation in intra-field interpolation and improving the image quality of the image after interpolation, and the program thereof An object is to provide an apparatus.

The present invention relates to a sequential scanning conversion apparatus that interpolates scanning lines of an interlaced scanning image and converts the scanning line into a sequential scanning image.
A pixel range setting means for sequentially specifying pixels to be interpolated in the interlaced scanning image, and sequentially setting a predetermined pixel range for the pixels to be interpolated around the specified pixel to be interpolated; and the pixel range Is set, a plurality of interpolation direction setting means for setting a plurality of interpolation directions corresponding to the pixel to be interpolated in this pixel range, and a difference value between pixels existing in each of the plurality of interpolation directions is detected, An interpolation direction detecting means for detecting an interpolation direction most correlated with the pixel to be interpolated, and an average of the values of the pixels in the set pixel range located on the interpolation direction detected by the interpolation direction detecting means A gist is provided with a replacement unit that obtains a value and replaces the pixel value to be interpolated with the average value to obtain a pixel corresponding to the sequentially scanned image.

  As described above, according to the present invention, since the direction that has the most influence on the interpolation element is obtained and the interpolation based on this direction is performed, the effect that accurate interpolation can be performed in the intra-field interpolation can be obtained.

<Embodiment 1>
In the present embodiment, processing for obtaining an interpolation pixel by computer processing will be described.

  FIG. 1 is a schematic configuration diagram of a progressive scanning conversion apparatus according to the present embodiment. In the present embodiment, a configuration implemented by software (program) will be described. As shown in FIG. 1, the scanning line interpolation method of the present embodiment interpolates an interlaced video signal. In the frame memory 1 for storing the video signal, an interpolation area setting unit 3 (also referred to as pixel range setting means) that determines an interpolation element area based on an interpolation element (also simply referred to as a pixel), and an interpolation element Interpolation direction dependency amount calculation unit 4 (interpolation) for determining the dependency amount of the direction Di with respect to the interpolation element from the direction (D1, D2,...) Preset in the memory 8 and the coefficient ki of the memory 6 Interpolation that determines the optimum interpolation direction from the results obtained by collecting the dependency amounts (D1, D2,...) Obtained by the interpolation direction dependency amount calculation unit 4 and the direction setting means and the difference value calculation means. A direction determining unit 5 (also referred to as an interpolation direction detecting unit), and an interpolation processing unit 7 (replacement unit) that uses the pixel based on the interpolation direction determined by the interpolation direction determining unit 5 and the average of the pixels as the value of the interpolation element of the frame memory 1 ) And . The image data in the frame memory 1 is displayed on the display unit 2.

  As the interpolation direction of the memory 8, as shown in FIG. 2, angles (D1, D2,...) With respect to the pixel (IN) of the interpolation element are stored.

  Further, the memory 6 stores the distance from the interpolation element to the peripheral element and the coefficient in association with each other.

  The progressive scan conversion apparatus configured as described above will be described below. FIG. 3 is an explanatory diagram for defining peripheral elements (vertical direction and oblique direction) of the interpolation element E when E is an interpolation element to be obtained. 3 uses E as an interpolation element when there are peripheral elements A, B, C... I in the interpolation line.

  FIG. 4 is an explanatory diagram of directions with respect to peripheral elements when the direction D1 is set with respect to the interpolation element E. FIG.

  In the present embodiment, as shown in FIGS. 5, 6, and 7, each pixel of a pixel line (a line that is not interpolated) is defined.

  For example, as shown in FIG. 5A, a symbol a is added to the upper left pixels of the pixels A, B, C,... I of the interpolation line to form pixels Aa, Ba,. .

  Further, as shown in FIG. 5 (b), a symbol b is added to the pixels A, B, C,... I immediately above the interpolation line to add pixels Ab, Bb,. To do.

  Further, as shown in FIG. 6A, a symbol c is added to the pixels on the upper right of the pixels A, B, C,... I of the interpolation line to form pixels Ac, Bc,. .

  Further, as shown in FIG. 6B, a symbol d is added to the lower left pixels of the pixels A, B, C,... I of the interpolation line to form pixels Ad, Bd,. .

  Further, as shown in FIG. 7A, a symbol e is added to pixels A, B, C,... I immediately below the interpolation line to form pixels Ae, Be,. .

  Further, as shown in FIG. 7B, the symbol A is added to the lower right pixel of the pixels A, B, C,... I of the interpolation line, and the pixels Af, Bf,. To do.

  FIG. 8 is a flowchart for explaining the operation of the present embodiment. The interpolation area setting unit 3 shown in FIG. 1 defines the interpolation element E of the interpolation line in the frame memory 1 (S1). In the present embodiment, an area of row jd and column ic is set as an interpolation element E (see FIG. 4).

  Next, the interpolation area setting unit 3 defines an area Qi (area including peripheral elements) centering on the interpolation element E (S2). Next, a region Pi for horizontal and vertical 5 pixels centering on the interpolation element E is defined on the region Qi (S3).

  Then, the interpolation direction dependence amount calculation unit 4 reads the direction Di (D1, D2,..., Also referred to as the first interpolation direction) in the memory 8 (S4), and the direction D1 based on the direction Di is determined as the region Qi. (S5).

  For example, in the direction D1 (angle), as shown in FIG. 4, the direction of D1A for the peripheral element A, the direction of D1B for the peripheral element B, and the direction of D1C for the peripheral element C For the peripheral element I, the direction of D1I is defined (the direction of nine components: also referred to as the second interpolation direction). The elements in these directions are diagonal three pixels.

  That is, D1A is (ja, ia), (jb, ib), (jc, ic), D1B is (ja, ib), (jb, ic), (jc, id), ... D1F is , (Jc, ic), (jd, id), (je, ie)... D1I is defined as (je, ic), (jf, id), (jg, ie).

  Then, the interpolation direction dependency amount calculation unit 4 calculates a dependency amount of Di (D1) with respect to the interpolation element E (S6). The calculation of the dependency amount of D1 will be described later in detail.

  Next, the interpolation direction dependency amount calculation unit 4 determines whether or not the direction Di is D5 (S7). If it is determined in step S7 that the direction is not D5, the direction Di is updated and the process returns to step S6 (S8).

  That is, the dependence amount of each direction with respect to E is obtained in the order of D1, D2, D3, D4, and D5.

  When it is determined in step S7 that the direction Di is the last direction D5, the interpolation direction determination unit 5 obtains a dependency amount in the direction having the strongest correlation among the dependency amounts in these directions (S9).

  Then, the interpolation direction determination unit 5 determines the direction having the dependency amount of the direction with the strongest correlation as the interpolation direction, and passes the pixel value in the interpolation direction to the interpolation processing unit 7 (S10).

  The interpolation processing unit 7 performs an interpolation process using the average of the pixel value (luminance) and the pixel value (luminance) in the direction obtained by the interpolation direction determination unit 5 as the value of the interpolation element E (S11). For example, when it is determined that D1 is the minimum (D1 has the strongest correlation), the interpolation element E is the average of the pixel Ea and the pixel Ef.

  Next, the interpolation processing unit 7 determines whether or not the interpolation elements (pixels) of all the interpolation lines have been interpolated (S12), and if not interpolated, updates to the next interpolation element (pixel). (S13), the process is returned to step S1, and the above process is performed.

  FIG. 9 is a flowchart for obtaining the dependency amount of the interpolation direction Di. In the present embodiment, the interpolation direction D1 will be described as a representative. The interpolation direction dependency amount calculation unit 4 has a peripheral element Mi (A (in this embodiment) A () ({ja, ia), (jb, ib), (jc, ic)}) with respect to the interpolation element E. jb, ib)) is read (S20).

  Then, the pixel value (luminance or gradation) of the upper left pixel ai (Aa in FIG. 5, Ba for B,... Ia for I) in FIG. 5 and the lower right pixel fi (FIG. 5). 7 (b) reads the pixel value (luminance or gradation) of Af if A, Bf if B,... If if (I) (S21). Next, the difference (difference value) between the pixel values (luminance or gradation) of the upper left pixel ai and the lower right pixel fi of the peripheral element Mi is obtained (S22a). Next, a distance Li from the interpolation element to the peripheral element Mi is obtained (S22b), a coefficient ki corresponding to the distance Li is selected from the memory 6 (S22c), and the coefficient ki is associated with the pixel value difference. And stored in a memory (not shown) (S23). Here, it supplements about the method of calculating | requiring the distance Li.

  For example, the distance from the interpolation element E to the peripheral elements (A to I) is obtained, and K0 to K3 (coefficients k0 to k3) are calculated. In this calculation, the distance between E and A is 3 pixels (or √5), the distance between E and B is 2 pixels, the distance between E and C is 3 pixels (or √5), the distance between E and D is 1 pixel, The distance between E and E is calculated as 0 pixel.

  That is, the difference values of the interpolation directions D1A, D1B,... D1I and the coefficient ki of the surrounding element Mi for the interpolation element are obtained.

  As the coefficient ki, for example, it is desirable to select k0 when the distance difference is the smallest and k3 when the distance difference is the smallest.

  Next, it is determined whether or not the difference value between ai and fi in the last direction (D1 = D1I) and the coefficient ki of the distance Li to the peripheral element with respect to the interpolation element have been obtained (S24). Updates D1i in the next direction (S25). For example, the direction D1i is updated to D1B, D1C,. Next, the updated D1i is set, and the process returns to step S21.

  On the other hand, in step S24, when it is determined that the difference value between the ai and fi in the last direction (D1 = D1I) and the coefficient ki are obtained, the above-described dependency amount is obtained by multiplying the coefficient and the difference value. An operation is performed (S27).

  That is, the dependence amount with respect to E in the D1 direction is obtained as shown in the following equation. In the present embodiment, the coefficient ki applied in each direction is represented by a predetermined formula.

[Equation 1]
D1 = K3 * D1A + K2 * D1B + K3 * D1C
+ K1 * D1D + K0 * D1E + K1 * D1F
+ K3 * D1G + K2 * D1H + K3 * D1I
= K3 | Aa-Af | + K2 | Ba-Bf | + K3 | Ca-Cf |
+ K1 | Da-Df | + K0 | Ea-Ef | + K1 | Fa-Ff |
+ K3 | Ga-Gf | + K2 | Ha-Hf | + k3 | Ia-If |
Asking.

The direction D2 is
[Equation 2]
D2 = K3 * D2A + K2 * D2B + K3 * D2C
+ K1 * D2D + K0 * D2E + K1 * D2F
+ K3 * D2G + K2 * D2H + K3 * D2I
= K3 | (Aa + Ab) / 2- (Ae + Af) / 2 |
+ K2 | (Ba + Bb) / 2- (Be + Bf) / 2 | + K3 | (Ca + Cb) / 2- (Ce + Cf) / 2 |
+ K1 | (Da + Db) / 2- (De + Df) / 2 | + K0 | (Ea + Eb) / 2- (Ee + Ef) / 2 |
+ K1 | (Fa + Fb) / 2- (Fe + Ff) / 2 |
+ K3 | (Ga + Gb) / 2- (Ge + Gf) / 2 | + K2 | (Ha + Hb) / 2- (He + Hf) / 2 |
+ K3 | (Ia + Ib) / 2- (Ie + If) / 2 |
The direction D3 is
[Equation 3]
D3 = K3 * D3A + K2 * D3B + K3 * D3C
+ K1 * D3D + K0 * D3E + K1 * D3F
+ K3 * D3G + K2 * D3H + K3 * D3I
= K3 | Ab-Ae | + K2 | Bb-Be | + K3 | Cb-Ce |
+ K1 | Db-De | + K0 | Eb-Ee | + K1 | Fb-Fe |
+ K3 | Gb-Ge | + K2 | Hb-He | + k3 | Ib-Ie |
Direction D4 is
[Equation 4]
D4 = K3 * D4A + K2 * D4B + K3 * D4C
+ K1 * D4D + K0 * D4E + K1 * D4F
+ K3 * D4G + K2 * D4H + K3 * D4I
= K3 | (Ab + Ac) / 2- (Ad + Ae) / 2 | + K2 | (Bb + Bc) / 2- (Bd + Be) / 2 |
+ K3 | (Cb + Cc) / 2- (Cd + Ce) / 2 |
+ K3 | (Db + Dc) / 2- (Dd + De) / 2 | + K0 | (Eb + Ec) / 2- (Ed + Ee) / 2 |
+ K1 | (Fb + Fc) / 2- (Fd + Fe) / 2 |
+ K3 | (Gb + Gc) / 2- (Gd + Ge) / 2 | + K2 | (Hb + Hc) / 2- (Hd + He) / 2 |
+ K3 | (Ib + Ic) / 2- (Id + Ie) / 2 |
Direction D5 is
[Equation 5]
D5 = K3 * D5A + K2 * D5B + K3 * D5C
+ K1 * D5D + K0 * D5E + K1 * D5F
+ K3 * D5G + K2 * D5H + K3 * D5I
= K3 | Ac-Ad | + K2 | Bc-Bd | + K3 | Cc-Cd |
+ K1 | Dc-Dd | + K0 | Ec-Ed | + K1 | Fc-Fd |
+ K3 | Gc-Gd | + K2 | Hc-Hd | + K3 | Ic-Id |
Then, the interpolation direction determination unit obtains the minimum one of D1 to D5 and determines the interpolation direction. That is, the direction in which the product of the difference value and the coefficient is small is the direction in which the correlation is strong. For example, when D1 is the minimum, the interpolation pixel E is the average of the pixel Ea and the pixel Ef.

  FIG. 10 is a diagram for explaining a specific example of the intra-field interpolation processing in the diagonal line of one horizontal pixel by each processing unit of the first embodiment. This time, the interpolation direction is determined based on the luminance value of the pixel. If the luminance of the white circle pixel is 100 and the luminance of the black circle pixel is 0, the pixel of luminance 0 is Bf, Ce, De, Ec, Ed, Fb, Gb, Ha, and the others are luminance 100 (see FIG. 11). reference). The values of D1 to D5 at this time are obtained.

From the above formula,
[Equation 6]
D1 = K3 | Aa-Af | + K2 | Ba-Bf | + K3 | Ca-Cf |
+ K1 | Da-Df | + K0 | Ea-Ef | + K1 | Fa-Ff |
+ K3 | Ga-Gf | + K2 | Ha-Hf | + K3 | Ia-If |
= K3 | 100-100 | + K2 | 100-0 | + K3 | 100-100 |
+ K1 | 100-100 | + K0 | 100-100 | + K1 | 100-100 |
+ K3 | 100-100 | + K2 | 0-100 | + K3 | 100-100 |
= 200 * K2
Similarly, [Equation 7]
D2 = 100 * K1 + 100 * K2 + 100 * K3
D3 = 200 * K1 + 200 * K3
D4 = 100 * K1 + 100 * K3
D5 = 0
K0 to K3 are weighting coefficients ki depending on the distance from the center pixel E, and are positive values. When D1 to D5 are compared, D5 is minimized, and the interpolation pixel E is the average of the pixel Ec and the pixel Ed. That is, appropriate interpolation is performed even for diagonal lines of less than two horizontal pixels.

That is, the software according to the first embodiment stores the scanned image in the first memory, and in the sequential scanning conversion program for displaying the image of the interpolation element of the interpolation line in the interlaced scanning on the screen,
On the computer,
Means for storing in a second memory a number of first interpolation directions each having a different angle;
Means for storing a weighting coefficient in a third memory in association with a distance from the interpolation element to a peripheral element;
Means for sequentially setting the interpolation elements in an interpolation line when the scanned image is stored in the first memory;
Each time the interpolation element is set, the first interpolation direction of the second memory is sequentially read, and the second interpolation in the same direction as the read first interpolation direction with respect to peripheral elements of the interpolation element Means for sequentially setting directions with reference to the peripheral elements,
Means for sequentially extracting pixel values of upper and lower or upper left and lower right pixel lines of the peripheral elements in the second interpolation direction each time a second interpolation direction based on the first interpolation direction is set;
Each time the pixel value in the second interpolation direction based on the first interpolation direction is extracted, a difference value between both pixel values in the second interpolation direction is obtained, and the interpolation element and the peripheral element are obtained. Means for determining the distance to and selecting the weighting coefficient corresponding to the distance from the third memory, and multiplying the coefficient by the difference value in the second interpolation direction;
Means for reading the multiplication results of all the second interpolation directions based on the first interpolation direction, adding them, and obtaining the added value as a dependency amount of the first interpolation direction with respect to the interpolation element ,
Each time the dependency amounts of the first interpolation directions are obtained, the dependency amounts of the first interpolation directions are compared with each other, and the first interpolation direction having the smallest dependency amount is obtained. Means for optimal interpolation direction for the interpolation element;
In order to realize a function as means for converting the interpolation value of the first memory into the interpolation value by using the average of the pixel values of the upper, lower, lower left and lower right in the optimal interpolation direction as the interpolation value of the interpolation element It can be said that this is a technical idea of a progressive scan conversion program.

<Embodiment 2>
In the present embodiment, an example in which the above processing is configured by hardware will be described. FIG. 12 is a schematic configuration diagram of the progressive scan conversion apparatus of the present embodiment.

  As shown in FIG. 12, an input terminal 20 for inputting one line of video signal of one field, a filter 23 (pixel range setting means, interpolation direction setting means), and a direction-dependent amount calculation unit 25 (difference value calculation). Means, an interpolation direction detecting means), a comparison circuit 26, and a selector 28. The comparison circuit 26 and the selector 28 correspond to interpolation direction detection means and replacement means.

  Line memories 21 a, 21 b, and 21 c that are delayed by one line are connected in series to the input terminal 20, and the output of the line memory 21 c is connected to the filter 23. The filter 23 includes filters for extracting pixels based on directions D1 to D5 shown in FIG. 2 for each direction (D1 filter 23a, D2 filter 23b,... D5 filter 23e). The filter 23 includes video data of a line input to the input terminal 20, video data for one line delayed by one line by the line memory 21a, video data delayed by two lines after the line memory 21b, and line memory 21c. The subsequent video data delayed by three lines is input, and pixels based on the direction component shown in FIG. 2 are output.

  The D1 filter 23a includes a pixel value (luminance) of video data of a line input to the input terminal 20, a pixel value (luminance or gradation) of video data for one line delayed by one line by the line memory 21a, and a line The pixel value (luminance or gradation) of the video data delayed by 2 lines after the memory 21b and the pixel value (luminance) of video data delayed by 3 lines after the line memory 21c are input, and the peripheral interpolation pixel (A Extract the pixel of the component in the D1 direction for .about.I).

  The D2 filter 23b has a pixel value (luminance) of video data of a line input to the input terminal 20, a pixel value (luminance) of video data for one line delayed by one line by the line memory 21a, and the line memory 21b The pixel value (luminance) of the video data delayed by two lines and the pixel value (luminance) of the video data delayed by three lines after the line memory 21c are input, and D2 for the peripheral interpolation pixels (A to I) of the interpolation element Extract the pixel of the direction component.

  The D3 filter 23c includes a pixel value (luminance or gradation) of video data of a line input to the input terminal 20, a pixel value (luminance) of video data for one line delayed by one line by the line memory 21a, and a line The pixel value (luminance) of the video data delayed by two lines after the memory 21b and the pixel value (luminance) of video data delayed by three lines after the line memory 21c are input, and the peripheral interpolation pixels (A to I) of the interpolation element The pixel of the component of D3 direction about is extracted.

  The D4 filter 23d is a pixel value (luminance) of video data of a line input to the input terminal 20, a pixel value (luminance) of video data for one line delayed by one line by the line memory 21a, and the line memory 21b The pixel value (luminance) of the video data delayed by 2 lines and the pixel value (luminance) of the video data delayed by 3 lines after the line memory 21c are input, and D4 for the peripheral interpolation pixels (A to I) of the interpolation element is input. Extract the pixel of the direction component.

  The D5 filter 23e is a pixel value (luminance) of video data of a line input to the input terminal 20, a pixel value (luminance) of video data for one line delayed by one line by the line memory 21a, and the line memory 21b The pixel value (luminance) of the two-line delayed video data and the pixel value (luminance) of the three-line delayed video data after the line memory 21c are input, and D5 for the peripheral interpolation pixels (A to I) of the interpolation element Extract the pixel of the direction component.

  For example, the D1 filter 23a includes pixels Aa, Ba,... Ia at the upper left of the peripheral interpolation pixels shown in FIG. 5A and pixels Af, Bf at the lower right of the peripheral interpolation pixels shown in FIG. , Df, Ef, Gf, and Hf are extracted.

  In addition, the direction dependency amount calculation unit 25 includes a D1 dependency amount calculation unit 25a,..., A D5 dependency amount calculation unit 25e.

  Each dependency amount calculation unit includes a pixel value (luminance) of video data of a line input to the input terminal 20, a pixel value (luminance) of video data for one line delayed by one line by the line memory 21a, and a line The pixel value (luminance) of the video data delayed by two lines after the memory 21b and the pixel value (luminance) of video data delayed by three lines after the line memory 21c are inputted and extracted from the filter in the corresponding direction. A pixel value is input, a difference between the respective pixel values is obtained, and a coefficient ki is multiplied to obtain a dependency amount in each direction. This coefficient ki is stored in correspondence with the distance. For example, the distance between E and A is 3 pixels (or √5), the distance between E and B is 2 pixels, the distance between E and C is 3 pixels (or √5), the distance between E and D is 1 pixel, The distance of E is calculated as 0 pixel,..., And a coefficient ki corresponding to this distance is selected.

  The comparison circuit 26 compares the dependency amounts in the directions D1, D2,... D5 from the direction dependency amount calculation unit 25, determines the most dependent direction, and outputs the determined direction to the selector 28.

  The selector 28 selects the average of the pixel value and the pixel value having the most influence in the direction output from the comparison circuit 26, and outputs the selected pixel value as the pixel value of the interpolation element. This pixel value is the E value of the screen.

  FIG. 13 is a detailed configuration diagram of the filter 23. In the present embodiment, the D1 filter 23a will be described as an example.

  As shown in FIG. 13, the line between the line memory 21a and the line memory 21b is the pixel line a shown in FIG. 14, and the line between the line memory 21b and the line memory 21c is the pixel shown in FIG. The line b is followed by the pixel line c shown in FIG. 14 after the line memory 21c.

  The input terminal 20 is connected in the order of pixel delays 30a, 30b,... 30d, and the input terminal of the pixel delay 30a is connected to the output terminal 31a (for If in FIG. 7B). An output terminal 31b (for Hf in FIG. 7B) is connected between the delay 30a and the pixel delay 30b.

  Further, an output end 31c (for Gf in FIG. 7B) is connected between the pixel delay 30b and the pixel delay 30c, and an output end 31d is connected between the pixel delay 30c and the pixel delay 30d. Has been. Further, an output terminal 31e (Gd in FIG. 6B) is connected to the pixel delay 30d.

  On the other hand, the pixel delays 32a, 32b,... 32d are connected in order between the line memory 21a and the line memory 21b, and the output terminal 33a (Ff in FIG. 7B) is connected to the input terminal of the pixel delay 32a. And an output end 33b (for Ef in FIG. 7B) is connected between the pixel delay 32a and the pixel delay 32b.

  An output terminal 33c (for Df in FIG. 7B) and an output terminal 33d (for Ia in FIG. 5A) are connected between the pixel delay 32b and the pixel delay 32c. An output terminal 33e (for Ha in FIG. 5A) is connected to the pixel delay 32d. Further, an output end 33f (for Ga in FIG. 5A) is connected to the pixel delay 32d.

  The line memories 21b and 21c are connected in the order of pixel delays 34a, 34b,... 34d, and the output terminal 35a (for Cf in FIG. 7B) is connected to the input terminal of the pixel delay 34a. Are connected, and an output end 35b (for Bf in FIG. 7B) is connected between the pixel delay 34a and the pixel delay 34b.

  An output terminal 35c (for Af in FIG. 7B) and an output terminal 35d (for Fa in FIG. 5A) are connected between the pixel delay 34b and the pixel delay 34c, and the pixel delay 34c An output end 35e (for Ea in FIG. 5A) is connected to the pixel delay 34d. Furthermore, an output end 35f (for Da in FIG. 5A) is connected to the pixel delay 34d.

  Further, pixel delays 36a, 36b,... 36d are connected to the line memory 21c in this order, and an output terminal 37a (for Ca in FIG. 5B) is connected between the pixel delay 36b and the pixel delay 36c. Has been.

  An output terminal 37b (for Ba in FIG. 5B) is connected between the pixel delay 36c and the pixel delay 36d, and an output terminal 37c (Aa in FIG. 5A) is connected to the pixel delay 36d. Is connected. That is, pixels related to the interpolation element in the direction D1 (D1A, D1B,... D1i) in FIG. 4 are extracted.

  FIG. 15 is a schematic configuration diagram of the D1 dependency calculation unit 25a. As shown in FIG. 15, difference units 40a, 40b,... 40i, coefficient multipliers 41a, 41b,... 41i, an adder 43, an adder 44, and an averaging circuit 45 are provided. I have.

  The coefficient multiplier 41a (for k3) is connected to a differentiator (subtracter) 40a for obtaining a difference between the pixel values Aa and Af, and an output terminal is connected to the adder 43.

  The coefficient multiplier 41b (for k2) is connected to a differentiator (subtractor) 40b for obtaining a difference between the pixel values Ba and Bf, and an output terminal is connected to the adder 43.

  The coefficient multiplier 41c (for k3) is connected to a differentiator (subtracter) 40c for obtaining a difference between the pixel values Ca and Cf, and an output terminal is connected to the adder 43.

  The coefficient multiplier 41d (for k1) is connected to a differentiator (subtracter) 40d for obtaining a difference between the pixel values Da and Df, and an output terminal is connected to the adder 43.

  The coefficient multiplier 41e (for k0) is connected to a differentiator (subtractor) 40e for obtaining a difference between the pixel values Ea and Ef, and an output terminal is connected to the adder 43.

  The coefficient multiplier 41f (for k1) is connected to a differentiator (subtracter) 40f for obtaining a difference between the pixel values Fa and Ff, and an output terminal is connected to the adder 43.

  The coefficient multiplier 41g (for k3) is connected to a differentiator (subtracter) 40g for obtaining a difference between the pixel values Ga and Gf, and an output terminal is connected to the adder 43.

  The coefficient multiplier 41h (for k2) is connected to a differentiator (subtracter) 40h for obtaining a difference between the pixel values Ha and Hf, and an output terminal is connected to the adder 43.

  The coefficient multiplier 41i (for k3) is connected to a differentiator (subtracter) 40i for obtaining a difference between the pixel values Ia and If, and an output terminal is connected to the adder 43.

  Further, an adder 44 for adding the pixel value Ea and the pixel value Ef is connected to the subtractor 40e, and an averaging circuit 45 is connected to the adder 44. The output of the averaging circuit 45 is connected to the selector 28.

  That is, in the progressive scan conversion device configured as described above, the pixel value in the direction shown in FIG. 14 is extracted by the D1 filter 23a in the direction D1, and is output to the D1 dependency calculation unit 25a.

  That is, when the image data of the area shown in FIG. 3 is input for each line, the pixel data of the C line shown in FIG. 14 is converted into the pixel delays 36a... 36d (line delay) by the line memories 21a, 21b and 21c. C line). Since there are two pixel delays 36a and 36b on the C line, as shown in FIG. 5A, (ja, id) and (ja, ie) are not read, and the output of the pixel delay 36b. The pixel value Ca of (ja, ic) is output, the pixel value Ba of (ja, ib) is output by the pixel delay 36c, and the pixel value Aa of (ja, ia) is output by the pixel delay 36d.

  Also, the line memories 21a and 21b (2 line delay) input the pixel data of the b line shown in FIG. 14 to the pixel delays 34a... 34d (referred to as the b line). Since the input end of the pixel delay 34a is used as the output end for the b line, the pixel value Cf of (jc, ie) is output to the output end 35a. Then, the pixel value Bf of (jc, id) delayed by one pixel by the pixel delay 34a is output. In addition, the pixel value Af of (jc, ic) and the pixel value Fa are output by the pixel delay 34b. Then, the pixel value Ea of (jc, ib) is output by the pixel delay 34c, and the pixel value Da of (jc, ic) is output by the pixel delay 34d.

  On the other hand, the line memory 21a (1 line delay) inputs the pixel data of the a line shown in FIG. 14 to the pixel delays 32a to 32d (referred to as a line). In the a line, since the input end of the pixel delay 32a is used as the output end, the pixel value Ff of (jd, ie) is output to the output end 33a. Then, the pixel value Ef of (jd, id) delayed by one pixel by the pixel delay 32a is output. Further, the pixel value Df of (je, ic) and the pixel value Ia are output by the pixel delay 32b. Then, the pixel value Ha of (jd, ib) is output by the pixel delay 32c, and the pixel value Ga of (je, ia) is output by the pixel delay 32d.

  Also, pixel values If, Hf,... Gd are output from locations not via the memory line.

  Then, it is input to the D1-dependent amount calculation unit 25a shown in FIG. 15, and the upper right and lower left pixel values of the respective peripheral elements (A to I) are subtracted and multiplied by a predetermined count. The adder 43 is entered.

That is, as in the first embodiment,
[Equation 8]
D1 = K3 * D1A + K2 * D1B + K3 * D1C
+ K1 * D1D + K0 * D1E + K1 * D1F
+ K3 * D1G + K2 * D1H + K3 * D1I
= K3 | Aa-Af | + K2 | Ba-Bf | + K3 | Ca-Cf |
+ K1 | Da-Df | + K0 | Ea-Ef | + K1 | Fa-Ff |
+ K3 | Ga-Gf | + K2 | Ha-Hf | + k3 | Ia-If |
Will be asking.

The direction D2 is
[Equation 9]
D2 = K3 * D2A + K2 * D2B + K3 * D2C
+ K1 * D2D + K0 * D2E + K1 * D2F
+ K3 * D2G + K2 * D2H + K3 * D2I
= K3 | (Aa + Ab) / 2- (Ae + Af) / 2 |
+ K2 | (Ba + Bb) / 2- (Be + Bf) / 2 | + K3 | (Ca + Cb) / 2- (Ce + Cf) / 2 |
+ K1 | (Da + Db) / 2- (De + Df) / 2 | + K0 | (Ea + Eb) / 2- (Ee + Ef) / 2 |
+ K1 | (Fa + Fb) / 2- (Fe + Ff) / 2 |
+ K3 | (Ga + Gb) / 2- (Ge + Gf) / 2 | + K2 | (Ha + Hb) / 2- (He + Hf) / 2 |
+ K3 | (Ia + Ib) / 2- (Ie + If) / 2 |
The direction D3 is
[Equation 10]
D3 = K3 * D3A + K2 * D3B + K3 * D3C
+ K1 * D3D + K0 * D3E + K1 * D3F
+ K3 * D3G + K2 * D3H + K3 * D3I
= K3 | Ab-Ae | + K2 | Bb-Be | + K3 | Cb-Ce |
+ K1 | Db-De | + K0 | Eb-Ee | + K1 | Fb-Fe |
+ K3 | Gb-Ge | + K2 | Hb-He | + k3 | Ib-Ie |
Direction D4 is
[Equation 11]
D4 = K3 * D4A + K2 * D4B + K3 * D4C
+ K1 * D4D + K0 * D4E + K1 * D4F
+ K3 * D4G + K2 * D4H + K3 * D4I
= K3 | (Ab + Ac) / 2- (Ad + Ae) / 2 | + K2 | (Bb + Bc) / 2- (Bd + Be) / 2 |
+ K3 | (Cb + Cc) / 2- (Cd + Ce) / 2 |
+ K3 | (Db + Dc) / 2- (Dd + De) / 2 | + K0 | (Eb + Ec) / 2- (Ed + Ee) / 2 |
+ K1 | (Fb + Fc) / 2- (Fd + Fe) / 2 |
+ K3 | (Gb + Gc) / 2- (Gd + Ge) / 2 | + K2 | (Hb + Hc) / 2- (Hd + He) / 2 |
+ K3 | (Ib + Ic) / 2- (Id + Ie) / 2 |
Direction D5 is
[Equation 12]
D5 = K3 * D5A + K2 * D5B + K3 * D5C
+ K1 * D5D + K0 * D5E + K1 * D5F
+ K3 * D5G + K2 * D5H + K3 * D5I
= K3 | Ac-Ad | + K2 | Bc-Bd | + K3 | Cc-Cd |
+ K1 | Dc-Dd | + K0 | Ec-Ed | + K1 | Fc-Fd |
+ K3 | Gc-Gd | + K2 | Hc-Hd | + K3 | Ic-Id |
Then, the interpolation direction determination unit obtains the minimum one of D1 to D5 and determines the interpolation direction. For example, when D1 is the minimum, the interpolation pixel E is the average of the pixel Ea and the pixel Ef. That is, Embodiment 2 is an apparatus for sequential scan conversion in which a scan image is input and an image of an interpolation element of an interpolation line in interlaced scanning is displayed on a screen.
A line memory group in which line memories for delaying one horizontal scanning period are connected in series for the peripheral element lines of the interpolation element, and a plurality of delayed lines from the line memory group are provided for each interpolation direction of the interpolation element. A filter that inputs pixel values of a pixel line and sequentially extracts pixel values of upper, lower, upper left, and lower right in the pixel line of peripheral elements of the interpolation element by sequentially delaying the pixel values in accordance with the interpolation direction. And each of the interpolation directions, each of which reads the upper and lower or upper left and lower right pixel values of the pixel line of the peripheral element from the filter unit having the same interpolation direction, Find the difference values, multiply each of these difference values by a predetermined weighting coefficient, add the respective multiplication results, and add this to the interpolation element The interpolation direction dependency amount calculation unit that outputs the dependency amount of the interpolation direction is compared with the dependency amounts of the multiple interpolation directions from the interpolation direction dependency amount calculation unit, and the smallest interpolation direction is optimized for the interpolation element. Interpolation direction determination unit as an interpolation direction, and an average of the upper, lower, lower left, and lower right pixel values of the optimal interpolation direction as an interpolation value of the interpolation element, and a pixel that converts the interpolation element of the memory into the interpolation value It can be said that the technical idea is an apparatus for progressive scanning conversion including an interpolation unit.

  The present invention can be applied to a scanning line interpolation device that generates an interpolation signal in an interlaced television signal.

It is a schematic block diagram of the program of the progressive scanning conversion of Embodiment 1 of this invention. It is explanatory drawing explaining many interpolation directions of the Embodiment 1. FIG. FIG. 11 is an explanatory diagram for defining peripheral elements (vertical direction and diagonal direction) of the interpolation element E when E is used as an interpolation element. It is explanatory drawing of the direction with respect to a peripheral element at the time of setting it as the direction D1 with respect to the interpolation element E. FIG. It is explanatory drawing explaining the definition of each pixel in the pixel line with respect to a peripheral element. It is explanatory drawing explaining the definition of each pixel in the pixel line with respect to a peripheral element. It is explanatory drawing explaining the definition of each pixel in the pixel line with respect to a peripheral element. 3 is a flowchart for explaining the operation of the first embodiment. It is a flowchart explaining the dependence amount calculation of the direction of Di. 6 is an explanatory diagram for explaining a specific example of intra-field interpolation in a diagonal line of one horizontal pixel by each processing unit of the first embodiment. FIG. It is explanatory drawing explaining the definition of each pixel in the pixel line with respect to the peripheral element of FIG. It is a schematic block diagram of the progressive scanning conversion apparatus of Embodiment 2 of this invention. It is a detailed block diagram of the filter 23 in the same Embodiment 2. It is explanatory drawing explaining the direction with respect to the periphery element based on D1 direction of the same Embodiment 2. FIG. It is a schematic block diagram of the dependence amount calculation part 25a for D1 in the same Embodiment 2. FIG. It is explanatory drawing explaining the interpolation by the average value of an up-and-down line pixel. It is explanatory drawing of the interpolation using the correlation of the diagonal direction in a field.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Frame memory 3 Interpolation area setting part 4 Interpolation direction dependence amount calculation part 5 Interpolation direction determination part 7 Interpolation processing part 23 Filter 25 Direction dependence quantity calculation part 26 Comparison circuit 28 Selector

Claims (2)

In a sequential scanning conversion device that interpolates the scanning lines of the interlaced scanning image and converts it to a sequential scanning image,
Pixel range setting means for sequentially specifying the pixels to be interpolated in the interlaced scanning image, centering on the specified pixels to be interpolated, and sequentially setting a predetermined pixel range for the pixels to be interpolated;
Interpolation direction setting means for setting a plurality of interpolation directions corresponding to the pixels to be interpolated in the pixel range each time the pixel range is set;
An interpolation direction detecting means for detecting a difference value between pixels existing in each of the plurality of interpolation directions and detecting an interpolation direction most correlated with the pixel to be interpolated;
By obtaining an average value of the values of the pixels in the set pixel range located on the interpolation direction detected by the interpolation direction detection means, and substituting the value of the pixel to be interpolated with the average value Replacement means for obtaining pixels corresponding to the sequentially scanned image;
A sequential scanning conversion device characterized by comprising: The interpolation direction detecting means includes
Each time the difference value is calculated, the difference value is multiplied by a predetermined weighting coefficient and then added, and the addition result is compared with each other to detect the interpolation direction most correlated with the pixel to be interpolated. The progressive scan conversion apparatus according to claim 1, wherein:

Images