JP4120872B2 - Image enlarging / reducing apparatus, method, and program - Google Patents

Description

  The present invention relates to an image enlarging / reducing apparatus and method for enlarging / reducing an image represented by image data, for example, and a program for causing a computer to execute an image enlarging / reducing method.

  Conventionally, image data obtained by photoelectrically reading an image recorded on a photographic film or image data obtained by imaging a subject with an imaging device such as a digital camera or a camera-equipped mobile phone is used as the image data. Enlargement and reduction are performed to fit the size of the monitor to be reproduced. In particular, when sending image data obtained by taking an image of a subject with a camera-equipped mobile phone as an e-mail, a service is also provided that scales the image data according to the size of the liquid crystal monitor of the destination mobile phone. ing.

  Such enlargement / reduction of the image data is performed by interpolating new pixels according to the enlargement ratio between the pixels constituting the image represented by the image data. As methods for interpolating new pixels in this manner, various methods such as a linear interpolation method, a nearest neighbor method, a bilinear method, and a bicubic method are known.

However, when the image data is enlarged / reduced using only a single method, there is a problem that blur or shaggy occurs in an edge portion included in the image. For this reason, a method has been proposed in which an edge component included in an image is detected and an interpolation operation is performed by different processing between the edge portion and the non-edge portion (see Patent Document 1).
JP 2002-312020 A

  By the way, in the method described in Patent Document 1, the interpolation calculation is performed by different processing for the edge portion and the non-edge portion. However, since the calculation method is the known method described above, the edge portion is still blurred. There is a problem that it ends up. For example, when the profile shape of four pixels G1 to G4 adjacent in series is as shown in FIG. 13A, a steep edge exists between the pixel G2 and the pixel G3. Also in the case shown in FIG. 13B, an edge exists between the pixel G2 and the pixel G3. Here, for example, when the pixel value of the interpolation pixel between the pixels G2 and G3 is obtained using the linear interpolation method, as shown in FIGS. 13A and 13B, a straight line connecting the pixels G2 and G3 in the vicinity of the interpolation pixel. The upper value is the pixel value of the interpolated pixel. For this reason, although the edge exists between the pixels G2 and G3, the change in the pixel value near the edge becomes smooth, and as a result, the edge portion on the enlarged / reduced image is blurred. .

  In addition, various methods such as the linear interpolation method, nearest neighbor method, bilinear method, and bicubic method described above use the pixel values of the pixels that are two-dimensionally distributed in the vicinity of the new pixel to change the pixel value of the new pixel. Since it is calculated, the calculation takes a long time.

  The present invention has been made in view of the above circumstances, and an object thereof is to enlarge and reduce an image at high speed so as not to blur an edge portion.

An image scaling device according to the present invention is an image scaling device for scaling an image.
3 for each of the pixel values of the three pixel pairs by filtering each of the three pixel pairs including two adjacent pixels included in the four pixels adjacent in series on the image with a differential filter. Filtering means for obtaining two primary differences by obtaining two primary differences and performing filtering processing by the differential filter on two adjacent primary difference pairs in the three primary differences, respectively;
Detection means for detecting whether or not an edge exists between two central pixels of the four pixels adjacent in series based on a positive / negative relationship between the three primary differences and the two secondary differences When,
A boundary setting means for setting a predetermined boundary between the pixels when an edge exists between the two central pixels;
Determining means for determining on which side of the predetermined boundary a position of a new pixel interpolated between the two central pixels;
When the position of the new pixel is on one side of the predetermined boundary, an interpolation operation is performed using a pixel value of at least one pixel existing on the one side of the image, and the new pixel When the position of the new pixel is on the other side of the predetermined boundary, an interpolation operation is performed using the pixel value of at least one pixel existing on the other side of the image. Performing interpolation calculation means for calculating a pixel value of the new pixel and calculating a pixel value of the new pixel by a predetermined interpolation calculation when there is no edge between the two central pixels;
The filtering process, the detection of whether the edge exists, the setting of the predetermined boundary when the edge exists, the determination and the calculation of the pixel value of the new pixel, and the case where the edge does not exist By calculating the pixel value of the new pixel in one direction on the image, an image that is enlarged or reduced in the one direction is obtained, and then the filtering process, whether the edge exists or not is obtained. Detection, setting of the predetermined boundary when the edge exists, determination and calculation of the pixel value of the new pixel, and calculation of the pixel value of the new pixel when the edge does not exist, The filtering means, the detecting means, and the boundary setting means so as to obtain an image that is also enlarged and reduced in the other direction by performing in the other direction above. Is characterized in that a control means for controlling said determining means and said interpolation calculation means.

  The “difference filter” includes not only a filter for obtaining a simple difference between two adjacent pixels but also a filter capable of obtaining a weighted difference. Specifically, a filter with an even number of taps, for example, a 2-tap filter with a filter value of (−1, 1) can be used.

  As the “predetermined boundary”, a line that bisects the distance between the two central pixels can be used.

  In addition, as the predetermined interpolation calculation when no edge exists between the two central pixels, various known methods such as a linear interpolation method, a nearest neighbor method, a bilinear method, and a bicubic method can be used. In particular, it is preferable to apply the interpolation calculation only in the one-dimensional direction.

In the image enlargement / reduction apparatus according to the present invention, the filtering means performs filtering processing by the difference filter on the center two pixels of the four pixels adjacent in series to obtain the center two pixels. As a means for further obtaining the difference between pixel values,
Based on the determination result of whether or not the absolute value of the difference between the pixel values of the two central pixels is greater than or equal to a predetermined threshold value, the detection means detects the center of the four pixels adjacent in series. It may be a means for detecting whether or not an edge exists between two pixels.

An image scaling method according to the present invention is an image scaling method for scaling an image.
3 for each of the pixel values of the three pixel pairs by filtering each of the three pixel pairs including two adjacent pixels included in the four pixels adjacent in series on the image with a differential filter. Two primary differences are obtained by filtering each of the two adjacent primary difference pairs in the three primary differences with the filtering by the difference filter,
Based on the positive / negative relationship between the three primary differences and the two secondary differences, it is detected whether there is an edge between the two central pixels of the four pixels adjacent in series,
When an edge exists between the two central pixels, a predetermined boundary is set between the pixels,
Determining which side of the predetermined boundary is the position of the new pixel interpolated between the central two pixels;
When the position of the new pixel is on one side of the predetermined boundary, an interpolation operation is performed using a pixel value of at least one pixel existing on the one side of the image, and the new pixel When the position of the new pixel is on the other side of the predetermined boundary, an interpolation operation is performed using the pixel value of at least one pixel existing on the other side of the image. To calculate a pixel value of the new pixel, and when no edge exists between the two central pixels, calculate a pixel value of the new pixel by a predetermined interpolation operation,
The filtering process, the detection of whether the edge exists, the setting of the predetermined boundary when the edge exists, the determination and the calculation of the pixel value of the new pixel, and the case where the edge does not exist By calculating the pixel value of the new pixel in one direction on the image, an image that is enlarged or reduced in the one direction is obtained, and then the filtering process, whether the edge exists or not is obtained. Detection, setting of the predetermined boundary when the edge exists, determination and calculation of the pixel value of the new pixel, and calculation of the pixel value of the new pixel when the edge does not exist, By performing in the other direction above, an image that is also enlarged and reduced in the other direction is obtained.

  The image enlargement / reduction method according to the present invention may be provided as a program for causing a computer to execute the method.

  According to the present invention, in one direction on the image, a filtering process using a difference filter is performed on each of three pixel pairs including two adjacent pixels included in four pixels adjacent in series on the image. Thus, three primary differences are obtained for each of the pixel values of the three pixel pairs. Further, filtering processing using a difference filter is performed on two adjacent primary difference pairs in the three primary differences to obtain two secondary differences. Here, the profile of the pixel values of four pixels adjacent in series can be classified according to the positive and negative values of the three primary differences and the two secondary differences, and the three primary differences and the two secondary differences can be classified. If a positive / negative relationship is seen, it can be determined whether or not an edge exists between two central pixels of four pixels adjacent in series. Therefore, based on the positive / negative relationship between the three primary differences and the two secondary differences, by detecting whether there is an edge between the two central pixels of the four pixels adjacent in series, It is possible to detect whether there is an edge between pixels, not on the pixels of the image.

  When an edge exists between the two central pixels, a predetermined boundary is set between the pixels, and it is determined which side of the predetermined boundary the new pixel is located. When the position of the new pixel is on one side of the predetermined boundary, an interpolation operation is performed using the pixel value of at least one pixel existing on one side of the image, and the pixel value of the new pixel Is calculated. On the other hand, when the position of the new pixel is on the other side of the predetermined boundary, the pixel value of the new pixel is obtained by performing an interpolation operation using the pixel value of at least one pixel existing on the other side of the image. Is calculated. For this reason, the pixel value of the new pixel is not affected by the pixel values of the pixels adjacent to the new pixel, and reflects only the pixel value of the pixel on one side. Therefore, as shown in FIGS. 13A and 13B, compared to the case where the pixel value of the new pixel is calculated using the pixel values of the pixels adjacent to the pixel value of the new pixel, The pixel value of a new pixel can be calculated so as to reduce blur. If there is no edge between the two central pixels, the pixel value of a new pixel is calculated by a predetermined interpolation calculation.

  Then, after calculating a pixel value of a new pixel in one direction of the image to obtain an image enlarged or reduced in one direction, a pixel value of a new pixel is calculated in the other direction in the same way, An image scaled in both directions is obtained.

  As described above, in this embodiment, since only the difference is calculated when detecting whether or not an edge exists, it is detected whether or not the edge exists at high speed by a simple calculation. be able to. In addition, since the pixel value of a new pixel is calculated so as to reduce edge blurring, an enlarged / reduced image with little edge blurring can be obtained.

  Also, when it is determined that an edge exists, if the absolute value of the difference between the pixel values of the two central pixels is small, the probability that the edge exists there is small, but if the absolute value of the difference is large, It can be assumed that there is a true edge there. Accordingly, based on the determination result of whether or not the absolute value of the difference between the pixel values of the two central pixels is equal to or greater than a predetermined threshold value, the distance between the two central pixels of the four pixels adjacent in series is determined. By detecting whether or not there is an edge, it is possible to reliably detect whether or not there is an edge between pixels, not on the pixels of the image.

  If there is an edge between the pixels of the image, the image is scaled in one direction, that is, the pixel value of the new pixel is calculated, and then the pixel value of the new pixel is calculated. Therefore, the pixel value of a new pixel can be calculated at a higher speed than the case where the pixel value of a new pixel is calculated by a single calculation in two dimensions.

  Hereinafter, embodiments of the present invention will be described. FIG. 1 is a schematic block diagram showing the configuration of an image enlargement / reduction apparatus according to an embodiment of the present invention. As shown in FIG. 1, the image enlargement / reduction apparatus according to the embodiment of the present invention calculates the pixel values of the input unit 1, the edge determination unit 2, and the interpolated pixels that accept input of the image data S0 and the enlargement factor K of the image data S0. And a control unit 4 that controls the operation of the input unit 1, the edge determination unit 2, and the interpolation calculation unit 3.

  Here, the image represented by the image data S0 (hereinafter, the reference symbol S0 is also used for the image) is composed of two-dimensionally arranged pixels as shown in FIG. In the embodiment, the x direction and the y direction are defined as shown in FIG.

  The edge determination unit 2 includes a filtering unit 2A and a determination unit 2B.

  The filtering unit 2A performs the filtering process as follows. FIG. 3 is a diagram for explaining the filtering process performed in the filtering unit 2A. The filtering unit 2A is adjacent to each of two adjacent pixels G2 and G3 between which the interpolation pixel P is located and each of the two pixels G2 and G3 for each column in the x direction and the y direction of the image S0. Differences between three pixel pairs G1 and G4, two pixel pairs G1 and G2, two pixel pairs G2 and G3, and two pixel pairs G3 and G4, which are adjacent to each other in the four pixels G1 to G4 that are adjacent to each other and that are composed of two pixels G1 and G4. Filtering processing by a filter is performed, and the differences between the pixel values of the pixel pairs G1, G2, the pixel pairs G2, G3, and the pixel pairs G3, G4 are calculated as primary differences d1, d2, d3.

  FIG. 4 is a diagram illustrating an example of the difference filter. As shown in FIG. 4, the differential filter used in the present embodiment is a 2-tap filter having a filter value of (−1, 1). Note that the difference filter is not limited to this, and a filter having a filter value for obtaining a weighted difference between two pixel values or a filter having an even number of taps of two or more taps may be used.

  Subsequently, the filtering unit 2A similarly performs filtering processing using a difference filter on two adjacent primary difference pairs d1, d2 and primary difference pairs d2, d3 in the three primary differences d1, d2, d3. The difference between the primary difference pairs d1, d2 and d2, d3 is calculated as secondary differences d4, d5.

  Further, the filtering unit 2A performs a filtering process on the central two pixels G2 and G3 of the four pixels G1 to G4 adjacent in series, and calculates a difference d0 (= d2) between the pixel values of the pixels G2 and G3. To do. The primary difference d2 may be used as the difference d0 as it is.

  The determination unit 2B determines whether or not an edge exists between two adjacent pixels G2 and G3 based on the positive / negative relationship between the primary differences d1, d2, and d3 and the secondary differences d4 and d5 ( First determination).

  FIGS. 5 to 7 are tables showing the relationship between the positive / negative of the primary differences d1, d2, d3 and the secondary differences d4, d5 of the four pixels adjacent in series and the profile shape of the four pixels adjacent in series. It is. There are a total of 18 combinations of positive and negative of the primary differences d1, d2, d3 and the secondary differences d4, d5 of four pixels adjacent in series. There are two combinations of edge 1 and edge 2 shown in FIG. 5 in which an edge exists between two adjacent pixels G2 and G3. Edge 1 is a right rising edge such that (d1, d2, d3, d4, d5) = (+, +, +, +, −) and (d1, d2, d3, d4, d5) = (−, −, − ,-, +). Edge 2 is a downwardly convex right rising edge such that (d1, d2, d3, d4, d5) = (+, +, +, +, +), (d1, d2, d3, d4, d5) = (+ , +, +,-,-) Upward convex right rising edge, (d1, d2, d3, d4, d5) = (-,-,-, +, +) downward convex left rising There are four types of edges, that is, upwardly convex left rising edges where (d1, d2, d3, d4, d5) = (−, −, −, −, −).

  The determination unit 2B stores the tables shown in FIGS. 5 to 7, and the positive / negative relationship between the primary differences d1, d2, d3 and the secondary differences d4, d5 of four pixels adjacent in series is shown in FIG. When the relationship between the edge 1 and the edge 2 shown in FIG. 2 is established, it is determined that an edge exists between two adjacent pixels G2 and G3. Further, when the positive / negative relationship of the primary differences d1, d2, d3 and the secondary differences d4, d5 of the four pixels adjacent in series is the peak, valley, and other relationships shown in FIGS. It is determined that no edge exists between two adjacent pixels G2 and G3.

  Further, when it is determined by the first determination that an edge is present, the determination unit 2B determines whether or not the absolute value of the difference d0 between the pixel values of the pixels G2 and G3 is equal to or greater than a predetermined threshold value Th1. (Second determination) When the second determination is affirmed, it is determined that an edge really exists. On the other hand, if the second determination is negative, it is determined that there is no edge there. In this way, the second determination is performed even if it is determined that there is an edge in the first determination, for example, as shown in FIG. This is to prevent the noise from being emphasized when the interpolation calculation unit 3 performs an interpolation calculation suitable for the edge portion as will be described later, assuming that the edge is an edge. When the determination is made in this way, the edge determination unit 2 outputs the determination result to the interpolation calculation unit 3.

  The interpolation calculation unit 3 includes a boundary setting unit 3A, a determination unit 3B, and a calculation unit 3C. Based on the determination result of the edge determination unit 2, an edge between two adjacent pixels between which the interpolation pixel P is located is an edge. The calculation unit 3C calculates the pixel value of the interpolation pixel P by different calculations depending on whether it is determined to be present or not determined to be an edge. Specifically, when it is determined that the edge is not an edge, the calculation unit 3C calculates the pixel value of the interpolation pixel P using the bicubic method.

  Here, the bicubic method is a method of cubic interpolation, and is a method of obtaining the pixel value of the interpolation pixel P using 16 pixels in the vicinity of the interpolation pixel P. Hereinafter, the bicubic method will be described.

FIG. 9 is a diagram for explaining the bicubic method. As shown in FIG. 9, when the point P is set as the position of the interpolated pixel P, a pixel indicated by a black circle in FIG. 9 is referred to as a primary neighborhood, and a pixel indicated by a white circle is referred to as a secondary neighborhood. For each of the first-order neighborhood and the second-order neighborhood, as shown in the following equation (1), the distances dx and dy (simply indicated as d in the equation (1)) independently in the x direction and the y direction, respectively. The weights Wx and Wy are obtained, and finally the weight W = WxWy for the pixel is obtained.

For example, when the weights Wx, Wy, and W are calculated for the pixel (second order neighborhood) of (−1, −1) in FIG.

It becomes.

When W (i, j) is the weight of the pixel (i, j) and f (i, j) is the pixel value of the pixel (i, j), the pixel value f ′ (P) of the interpolation pixel P is

Can be calculated.

  In the present embodiment, the pixel value of the interpolation pixel P is calculated by applying the bicubic method only in the one-dimensional direction of the x direction or the y direction.

  On the other hand, for the portion determined to be an edge, the pixel value of the interpolation pixel P is obtained as follows. FIG. 10 is a diagram illustrating a profile of pixel values for a portion determined to be an edge. In FIG. 10, the horizontal direction is the direction in which the pixels are arranged, and the vertical direction is the direction that indicates the magnitude of the pixel value. When it is determined that an edge exists between two adjacent pixels G2 and G3, two pixels G1 to G4 adjacent in series obtained by adding two adjacent pixels G1 and G4 to the two pixels, respectively. The value profile is as shown in FIG. 10 (a) or FIG. 10 (b).

  When the profile has the step edge shape shown in FIG. 10A, the boundary setting unit 3A sets the middle line M (see the alternate long and short dash line) that bisects the distance between the pixels G2 and G3 in the pixel arrangement direction as a boundary. To do. Then, the determination unit 3B determines whether the interpolation pixel P exists on the right side or the left side of the middle line M, and when the interpolation pixel (referred to as P1) exists on the right side of the middle line M, the calculation unit 3C Calculates the value on the extension of the straight line connecting the pixels G3 and G4 as the pixel value of the interpolation pixel P1. When an interpolation pixel (P2) is present on the left side of the middle line M, the calculation unit 3C calculates a value on an extension line of a straight line connecting the pixels G1 and G2 as the pixel value of the interpolation pixel P2.

  When the profile has the edge shape shown in FIG. 10B, the boundary setting unit 3A sets an intersection C between a straight line extending from the pixels G1 and G2 and a straight line extending from the pixels G3 and G4 as a boundary. To do. Then, the determination unit 3B determines whether the interpolation pixel P exists on the right side or the left side of the intersection C. If the interpolation pixel P1 exists on the right side of the intersection C, the calculation unit 3C determines that the pixels G3, G4 Is calculated as the pixel value of the interpolation pixel P1. When the interpolation pixel P2 exists on the left side of the intersection C, the calculation unit 3C calculates the value on the extension line of the straight line connecting the pixels G1 and G2 as the pixel value of the interpolation pixel P2.

  Here, the pixel value of the interpolation pixel P is calculated using only the pixel values of two pixels, but the pixel values of three or more pixels may be used. In this case, it may be difficult to connect the pixels with a straight line. For this reason, the pixels may be connected by a curve defined by an arbitrary function such as a spline curve, and the value on the extension line of the curve may be the value of the interpolation pixel P.

  Here, in the following description, the calculation when it is determined to be an edge is referred to as a first interpolation calculation, and the calculation when it is determined not to be an edge is referred to as a second interpolation calculation.

  Next, processing performed in the present embodiment will be described. FIG. 11 is a flowchart showing processing performed in the present embodiment. In the present embodiment, it is assumed that the interpolation pixel P exists between the pixels of the image S0. First, the input unit 1 accepts input of the image data S0 to be enlarged and the enlargement factor K of the image data S0 (step S1). The direction for interpolating the pixels is first set in the x direction (step S2). Next, edge determination is performed on the first interpolation pixel P (for example, the pixel located at the upper left of the image represented by the enlarged image data S1 (hereinafter, the reference symbol S1 is also used for the image)) corresponding to the enlargement factor K. The filtering unit 2A of the unit 2 calculates the primary differences d1, d2, d3 and the secondary differences d4, d5 from the four pixels G1 to G4 adjacent in series with the interpolation pixel P positioned therebetween (step S3). Further, the filtering unit 2A calculates a difference d0 by performing a filtering process using a difference filter on the pixels G2 and G3 (step S4).

  Then, based on the positive / negative relationship between the primary differences d1, d2, d3 and the secondary differences d4, d5, the determination unit 2B has an edge between two adjacent pixels G2, G3 between which the interpolation pixel P is located. It is determined whether or not it exists (first determination, step S5). If step S5 is affirmed, the determination unit 2B determines whether or not the absolute value of the difference d0 is greater than or equal to the threshold value Th1 (second determination, step S6).

  If step S6 is affirmed, it is assumed that an edge exists between the two central pixels G2 and G3 of the four pixels G1 to G4 adjacent in series, and the interpolation calculation unit 3 performs the interpolation pixel by the first interpolation calculation. A pixel value of P is calculated (step S7). On the other hand, if step S5 and step S6 are negative, the interpolation calculation unit 3 calculates the pixel value of the interpolation pixel P by the second interpolation calculation, assuming that no edge exists between the two central pixels G2 and G3. (Step S8).

  FIG. 12 is a flowchart of the first interpolation calculation. First, the boundary setting unit 3A of the interpolation calculation unit 3 sets the middle line M or the intersection C between the two central pixels G2 and G3 as a boundary (step S21). Then, the determination unit 3B determines which side of the boundary the interpolation pixel P is on (step S22), and the calculation unit 3C performs an interpolation calculation using only the pixel on the side where the interpolation pixel P exists, thereby interpolating the pixel. A pixel value of P is calculated (step S23).

  Returning to FIG. 11, next, the control unit 4 determines whether or not the pixel values have been calculated for all the interpolation pixels P (step S9). If step S9 is negative, the interpolation pixel P for calculating the pixel value is set to the next. Is set to the interpolation pixel P (step S10), and the process returns to step S3.

  If step S9 is affirmed, it is determined whether or not pixel values have been calculated for all interpolation pixels P in the y direction (step S11). If step S11 is negative, the interpolation direction is set to the y direction (step S12), and the process returns to step S3. When step S11 is affirmed, the enlarged image data S1 including the interpolation pixel P is output (step S13), and the process is terminated.

  As described above, in this embodiment, when detecting whether or not an edge is present, only the difference d0, the primary differences d1, d2, and d3 and the secondary differences d4 and d5 are calculated as described above. Therefore, it is possible to detect whether or not an edge exists at high speed by a simple calculation.

  If there is an edge between the pixels of the image S0, it is determined on which side of the boundary the interpolation pixel P exists as shown in FIGS. 10 (a) and 10 (b), and the interpolation pixel P Since the pixel value of the interpolation pixel P is calculated using only the pixels on the side where the pixel exists, the pixel value of the interpolation pixel P is not affected by the pixel values of the pixels G2 and G3 adjacent to the interpolation pixel P. Only the pixel value on one side is reflected. Therefore, as shown in FIGS. 13A and 13B, the edge value of the interpolation pixel P is calculated as compared with the case of calculating the pixel value of the interpolation pixel P using the pixel values of the pixels G2 and G3 adjacent to the interpolation pixel P. The pixel value of the interpolated pixel P can be calculated so as to reduce the blur, thereby obtaining an enlarged / reduced image S1 with a reduced edge blur.

  When an edge exists between the pixels of the image S0, the image S0 is enlarged / reduced in one direction, that is, the pixel value of the interpolation pixel P is calculated and then enlarged / reduced in the other direction, that is, the pixel value of the interpolation pixel P. Therefore, the pixel value of the interpolation pixel P can be calculated at a higher speed than the case where the pixel value of the interpolation pixel P is calculated by one-time calculation in two dimensions.

  In the above embodiment, the primary differences d1, d2, d3 and the secondary differences d4, d5 are calculated by performing the filtering process using the difference filter shown in FIG. 4 on the adjacent pixels G1 to G4, and the primary difference d1. , D2, d3 and the secondary difference d4, d5, and whether or not an edge exists between the pixels G2 and G3 based on the determination result of whether or not the absolute value of the difference d0 is greater than or equal to the threshold Th1 Even if it is determined whether or not an edge exists between the pixels G2 and G3 based only on the positive / negative relationship of the primary differences d1, d2, d3 and the secondary differences d4, d5. Good.

  In the above embodiment, when it is determined that the edge is not an edge, the pixel value of the interpolation pixel P is calculated from the pixel values of 16 pixels near the interpolation pixel P (4 pixels in the one-dimensional direction) using the bicubic method. However, the pixel value of the interpolation pixel P may be calculated from the pixel values of nine pixels (three pixels in the one-dimensional direction) or four pixels (two pixels in the one-dimensional direction) near the interpolation pixel P. In addition, the pixel value of the interpolation pixel P is calculated by one-dimensional interpolation calculation in the x direction or the y direction, but the pixel of the interpolation pixel P is calculated by two-dimensional interpolation calculation only when it is determined that no edge exists. A value may be calculated. Furthermore, the pixel value of the interpolation pixel P may be calculated using not only the bicubic method but also a linear interpolation method, a nearest neighbor method, a bilinear method, or the like.

1 is a schematic block diagram showing the configuration of an image enlargement / reduction device according to an embodiment of the present invention. The figure which shows the pixel arrangement | sequence of the image represented by image data The figure for demonstrating the filtering process performed in a filtering part The figure which shows the example of difference filter Table showing the relationship between the positive and negative of primary differences d1, d2, d3 and secondary differences d4, d5 and the profile shapes of four pixels adjacent in series (part 1) Table showing the relationship between the positive and negative of primary differences d1, d2, d3 and secondary differences d4, d5 and the profile shapes of four pixels adjacent in series (part 2) Table showing the relationship between the positive and negative of primary differences d1, d2, d3 and secondary differences d4, d5 and the profile shapes of four pixels adjacent in series (part 3) The figure which shows the example of the profile shape determined as an edge, although the difference of the pixel value of two adjacent pixels is very slight Illustration for explaining the bicubic method The figure for demonstrating calculation of the pixel value of the interpolation pixel about the part determined to be an edge A flowchart showing processing performed in the present embodiment First interpolation calculation flowchart Diagram for explaining conventional interpolation calculation

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Input part 2 Edge determination part 2A Filtering part 2B Determination part 3 Interpolation operation part 3A Boundary setting part 3B Determination part 3C Operation part 4 Control part

Claims (4)

In an image enlargement / reduction device that enlarges / reduces an image,
3 for each of the pixel values of the three pixel pairs by filtering each of the three pixel pairs including two adjacent pixels included in the four pixels adjacent in series on the image with a differential filter. Filtering means for obtaining two primary differences by obtaining two primary differences and performing filtering processing by the differential filter on two adjacent primary difference pairs in the three primary differences, respectively;
Detection means for detecting whether or not an edge exists between two central pixels of the four pixels adjacent in series based on a positive / negative relationship between the three primary differences and the two secondary differences When,
A boundary setting means for setting a predetermined boundary between the pixels when an edge exists between the two central pixels;
Determining means for determining on which side of the predetermined boundary a position of a new pixel interpolated between the two central pixels;
When the position of the new pixel is on one side of the predetermined boundary, an interpolation operation is performed using a pixel value of at least one pixel existing on the one side of the image, and the new pixel When the position of the new pixel is on the other side of the predetermined boundary, an interpolation operation is performed using the pixel value of at least one pixel existing on the other side of the image. Performing interpolation calculation means for calculating a pixel value of the new pixel and calculating a pixel value of the new pixel by a predetermined interpolation calculation when there is no edge between the two central pixels;
The filtering process, the detection of whether the edge exists, the setting of the predetermined boundary when the edge exists, the determination and the calculation of the pixel value of the new pixel, and the case where the edge does not exist By calculating the pixel value of the new pixel in one direction on the image, an image that is enlarged or reduced in the one direction is obtained, and then the filtering process, whether the edge exists or not is obtained. Detection, setting of the predetermined boundary when the edge exists, determination and calculation of the pixel value of the new pixel, and calculation of the pixel value of the new pixel when the edge does not exist, The filtering means, the detecting means, and the boundary setting means so as to obtain an image that is also enlarged and reduced in the other direction by performing in the other direction above. Image scaling device characterized by a control means for controlling said determining means and said interpolation calculation means. The filtering means is means for further obtaining a difference between pixel values of the central two pixels by performing filtering processing by the differential filter on two central pixels of the four pixels adjacent in series.
The detection means is based on a determination result of whether or not an absolute value of a difference between pixel values of the two central pixels is equal to or greater than a predetermined threshold value, and the center of the four pixels adjacent in series is determined. 2. The image enlarging / reducing apparatus according to claim 1, wherein the image enlarging / reducing apparatus is means for detecting whether or not an edge exists between two pixels. In an image scaling method for scaling an image,
3 for each of the pixel values of the three pixel pairs by filtering each of the three pixel pairs including two adjacent pixels included in the four pixels adjacent in series on the image with a differential filter. Two primary differences are obtained by filtering each of the two adjacent primary difference pairs in the three primary differences with the filtering by the difference filter,
Based on the positive / negative relationship between the three primary differences and the two secondary differences, it is detected whether there is an edge between the two central pixels of the four pixels adjacent in series,
When an edge exists between the two central pixels, a predetermined boundary is set between the pixels,
Determining which side of the predetermined boundary is the position of the new pixel interpolated between the central two pixels;
When the position of the new pixel is on one side of the predetermined boundary, an interpolation operation is performed using a pixel value of at least one pixel existing on the one side of the image, and the new pixel When the position of the new pixel is on the other side of the predetermined boundary, an interpolation operation is performed using the pixel value of at least one pixel existing on the other side of the image. To calculate a pixel value of the new pixel, and when no edge exists between the two central pixels, calculate a pixel value of the new pixel by a predetermined interpolation operation,
The filtering process, the detection of whether the edge exists, the setting of the predetermined boundary when the edge exists, the determination and the calculation of the pixel value of the new pixel, and the case where the edge does not exist By calculating the pixel value of the new pixel in one direction on the image, an image that is enlarged or reduced in the one direction is obtained, and then the filtering process, whether the edge exists or not is obtained. Detection, setting of the predetermined boundary when the edge exists, determination and calculation of the pixel value of the new pixel, and calculation of the pixel value of the new pixel when the edge does not exist, An image enlarging / reducing method, characterized in that an image enlarged / reduced in the other direction is obtained by performing in the other direction above. In a program for causing a computer to execute an image scaling method for scaling an image,
3 for each of the pixel values of the three pixel pairs by filtering each of the three pixel pairs including two adjacent pixels included in the four pixels adjacent in series on the image with a differential filter. A procedure of obtaining two primary differences by obtaining two primary differences and applying filtering processing by the differential filter to two adjacent primary difference pairs in the three primary differences, respectively.
A procedure for detecting whether or not an edge exists between two central pixels of the four pixels adjacent in series based on a positive / negative relationship between the three primary differences and the two secondary differences; ,
A step of setting a predetermined boundary between the pixels when an edge exists between the two central pixels;
A procedure for determining on which side of the predetermined boundary a position of a new pixel to be interpolated between the central two pixels;
When the position of the new pixel is on one side of the predetermined boundary, an interpolation operation is performed using a pixel value of at least one pixel existing on the one side of the image, and the new pixel When the position of the new pixel is on the other side of the predetermined boundary, an interpolation operation is performed using the pixel value of at least one pixel existing on the other side of the image. To calculate the pixel value of the new pixel, and when no edge exists between the two central pixels, a procedure for calculating the pixel value of the new pixel by a predetermined interpolation operation;
The filtering process, the detection of whether the edge exists, the setting of the predetermined boundary when the edge exists, the determination and the calculation of the pixel value of the new pixel, and the case where the edge does not exist By calculating the pixel value of the new pixel in one direction on the image, an image that is enlarged or reduced in the one direction is obtained, and then the filtering process, whether the edge exists or not is obtained. Detection, setting of the predetermined boundary when the edge exists, determination and calculation of the pixel value of the new pixel, and calculation of the pixel value of the new pixel when the edge does not exist, And a procedure for obtaining an image that is also enlarged and reduced in the other direction by performing in the other direction.

Images