JP3026706B2 - Image processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image output apparatus such as a printer which enlarges and resizes input image information and outputs the image information, and communication between models having different resolutions to convert low-resolution information to high-resolution information. The present invention relates to an image processing device for converting resolution.

[0002]

2. Description of the Related Art Conventionally, various methods have been proposed as a method of converting resolution of inputted low-resolution information into high-resolution information. These methods are based on the type of the target image (for example, a multi-valued image having gradation information for each pixel, a binary image binarized by pseudo halftone, a binary image binarized by a fixed threshold). Image, character image, etc.). The image targeted by the present invention is a multi-valued image such as a natural image having gradation information for each pixel, but the conventional interpolation method uses the same pixel value closest to the interpolation point as shown in FIG. The closest interpolation method of arranging, and the four points surrounding the interpolation points as shown in FIG.
In general, bilinear interpolation or the like for determining the pixel value E by the following calculation based on the distances C and D) is used.

E = (1-i) (1-j) A + i (1-
j) B + j (1-i) C + ijD However, when the distance between pixels is 1, it is assumed that there is a distance i from the A in the horizontal direction and j in the vertical direction (i ≦ 1, j ≦
1). As another method, there is a high-density pixel restoration method described in Japanese Patent Application Laid-Open No. 55-112076, and this method enables adaptive interpolation processing from low-resolution information to high-resolution information.

[0004]

However, the above-mentioned prior art has the following drawbacks. That is,
The method of FIG. 17 has an advantage that the configuration is simple, but when the target image is used for a natural image or the like, the pixel value is determined for each block to be enlarged, so that the blocks are visually conspicuous and the image quality is low. Poor.

The method shown in FIG. 18 is a commonly used method for enlarging natural pixels. With this method, the image quality is averaged and smoothed,
The image quality becomes blurred at the edge portion or at the portion where sharp image quality is required. Furthermore, in the case of an image obtained by scanning a map or the like or a natural image including a character portion, important information may not be transmitted to a recipient due to blurring due to interpolation.

In the above-mentioned Japanese Patent Application Laid-Open No. 55-112076, the enlargement magnification is fixed to three times, and the processed image information becomes binary information of white and black. This is equivalent to binarization using a simple dither pattern. SUMMARY An advantage of some aspects of the invention is to provide an image processing apparatus that can easily convert a low-resolution image into a high-resolution image and obtain high-quality image quality.

[0007]

To achieve the above object, an image processing apparatus according to the present invention has the following arrangement. The low resolution target pixel is enlarged to a block of (N × M) pixels,
An image processing apparatus for converting low-resolution image to a high resolution image, the surrounding pixels of the low resolution pixel of interest and its
Detecting means for detecting a maximum value and a minimum value ; determining means for determining a distribution ratio for arranging the maximum value and the minimum value detected by the detecting means in the block; and a distribution ratio determined by the determining means. Depending on the maximum and minimum
Arrangement means for arranging values .

Preferably, the determining means includes a maximum value to be arranged in the block so as to preserve the density of the pixel of interest.
It is characterized in that the area ratio of the value and the minimum value and the number of pixels are determined. Further preferably, further comprising an interpolation means for interpolating low-resolution pixels, and a sorting means for sorting the pixels interpolated by the interpolation means, wherein the arranging means, according to the area ratio, and the number of pixels And arranging the maximum value and the minimum value on the pixels sorted by the sorting means.

[0009] More preferably, the arranging means is arranged in a block according to a magnitude relationship between the surrounding pixels and a ratio.
It is characterized in that a maximum value and a minimum value are arranged. More preferably, the arranging means calculates a corner having a maximum value among four corners around the pixel of interest, and calculates a maximum value and a minimum value in a block according to a ratio between pixels adjacent to sides sandwiching the corner.
It is characterized by arranging values .

More preferably, the arranging means arranges a maximum value and a minimum value in a block according to a ratio between upper and lower or left and right pixels of the target pixel.
In addition, smoothing is performed after the arrangement.

[0011]

In such a configuration, when the low-resolution target pixel is enlarged into a block of (N × M) pixels and the low-resolution image is converted into a high-resolution image, the low-resolution target pixel and the low-resolution target pixel are converted to a high-resolution image.
Detecting the maximum and minimum values from the surrounding pixels patron, the maximum value and the minimum in accordance with the detected maximum value and the minimum value determines the distribution ratio to place in the block, the determined distribution ratio
Acts like placing a value .

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A preferred embodiment according to the present invention will be described below in detail with reference to the drawings. It is efficient that the image processing apparatus according to the present invention is mainly provided inside an image output apparatus such as a printer. However, the image processing apparatus other than the image output apparatus may be incorporated as application software in a host computer. It is possible.

<First Embodiment> FIG. 1 is a block diagram showing a configuration of an image processing apparatus according to a first embodiment. In the figure, reference numeral 100 denotes an input terminal to which low-resolution image information (low-resolution information) is input. This low-resolution information is transmitted to the linear interpolation means 101, and the pixels between the original samplings are buried by the linear interpolation (bilinear interpolation) processing to create interpolation information N times vertically and M times horizontally. . Figure 1 for linear interpolation
7, the description is omitted. Reference numeral 102 denotes an edge creating unit, which creates an edge for each block of interpolation information of N pixels × M pixels centering on a target pixel ( _Ixy ) of low resolution information. The details of the edge creation will be described later.

Reference numeral 103 denotes an allocation ratio determining means for calculating the allocation ratio (a, where 0 ≦ a ≦ 1) in the synthesis of the created linear interpolation information and edge information.
This distribution ratio is also determined for each block of N pixels × M pixels. Details of the determination of the distribution ratio will be described later.
Thereafter, using the obtained distribution ratio (a), the edge information is multiplied by a in the multiplier 104, the linear interpolation information is multiplied by (1−a) in the multiplier 105, and then synthesized by the adder 106. You. Reference numeral 107 denotes an output terminal whose input image information is N
It is converted into information of × M times and output.

FIG. 2 shows the edge creating means 10 shown in FIG.
2 is a diagram showing a detailed configuration of FIG. This edge creating means 102 is a feature of the first embodiment. A portion surrounded by a broken line corresponds to the edge creating means 102. In the figure, reference numeral 110 denotes an input terminal for low resolution information, to which information indicated by 111 is input. At 111, the pixel of E corresponds to the pixel of interest, and the portion surrounded by the dashed line is the window for the neighboring pixels of the pixel of interest. Reference numeral 112 denotes an input terminal from the linear interpolation means 101, to which information shown in 113 is input.
In 113, a broken line indicates a block boundary centered on each pixel of low-resolution information, and a portion surrounded by a solid line is a block for the target pixel E. In addition, a triangle indicates a pixel of low resolution information, and a triangle indicates an interpolation pixel.

With respect to the low resolution information 111 input from the input terminal 110, the maximum value (MAX) and the minimum value (MIN) in the window are detected by the MAX / MIN detecting means 114. The detected MAX and MIN are sent to the arrangement pixel calculation means 115, and the number of pixels of each of the binary representative values for forming the edge is calculated. Assuming that the number of pixels for arranging the representative value of _MAX is DOT _MAX and the number of pixels for arranging the representative value of _MIN is DOT _MIN , the number of pixels is determined by the following equation.

DOT _MAX = (I _xy −MIN) × N × M /
(MAX-MIN) DOT _MIN = N × M-DOT _{MAX The} determined DOT, MAX and MIN are sent to the dot arrangement means 116. On the other hand, the input terminal 11
The block 113 centered on the target pixel E subjected to the linear interpolation input from 2 is sent to the sorting means 117, and the linearly interpolated pixels in the block are sorted in descending order of the pixel value. The sorting algorithm itself is not limited here.

MAX is assigned to the pixels in the block including the sorted E by the dot arrangement means 116 by DOT _MAX pixels in descending order of the pixel value, and MIN is assigned to the other pixels in the block. You. 118
Denotes an output terminal, and as shown in 119, information of a block in which the binarized representative value is assigned by MAX and MIN is output.

FIG. 3 is a diagram showing the above-described linear interpolation and edge creation. For simplicity, it is shown in a one-dimensional direction. In the figure, the symbol 〇 indicates the pixel value of a sampling point on a low resolution, and the symbol X indicates the pixel value of an interpolation point interpolating between them. In addition, a point indicated by a triangle located at the center is set as a target pixel. As shown in FIG. 3A, M
AX and MIN are detected respectively, and D
Calculate OT _MAX . Now, assuming that the enlarged pixel block centered on the target pixel is 5 pixels and DOT _MAX is 4 pixels, as shown in FIG. 2B, the 4 pixels having a large pixel value are equal to MAX, and the pixel value is One small pixel is equal to MIN.

FIG. 4 is a diagram showing an example of creating edge information. Here, the calculation formula of the above-described DOT calculation will be described. In the example shown in FIG.
It shall be converted to double high resolution information. Now, N = M =
8 is taken as an example. Also, as shown in FIG.
It is assumed that the target pixel has a pixel value of “80” arranged in the center. When the low-resolution pixel “80” is considered in terms of the high-resolution information, as shown in (b), the edge portion in which two values of 200 pixels and 20 pixels are included in a certain ratio And
It is assumed that the value has been rounded to a value of 80 for low resolution. That is, when density preservation is considered, assuming that the ratio of the high resolution 200 included in the pixels of the low resolution 80 is A and the ratio of 20 is B, 200 × A + 20 × B = 80.

Here, if the variables I _xy , MAX, and MIN are used, MAX × A + MIN × B = I _{xy Since} A + B = 1 is assumed, MAX × A + MIN × (1-A) = I _xy a = a _{(I xy -MIN) / (MAX} -MIN).

Now, in order to enlarge one low-resolution pixel to (N × M) pixels, the number of pixels for MAX is: DOT _MAX = (I _xy −MIN) × N × M / (MAX × M
IN). In this example, I _xy = 80, MAX = 20
Since 0, MIN = 20, and N = M = 8, when substituted into the above equation, the number of pixels DOT _{MAX at} which the MAX value is arranged
Is equivalent to 21 pixels. That is, of the 64 pixels of 8 × 8 times, the value of MAX of 200 is assigned to 21 pixels in descending order of pixel value, and MIN is assigned to the remaining 43 pixels.
Is substituted.

As described above, an edge is created in a block corresponding to a target pixel. By the edge creation processing described above, it is possible to create a smooth edge in the high resolution direction, that is, high resolution information, within the enlarged block. FIG. 5 is a diagram showing a detailed configuration of the distribution ratio determining means 103. In the figure, a portion surrounded by a broken line corresponds to the distribution ratio determining means 103. In the figure, reference numeral 121 denotes an input terminal, to which the low-resolution information shown by 122 is input. As described with reference to FIG. 2, the pixel E is set as a target pixel. Similarly to the above-described edge creating means 102, the MAX and MIN detecting means 123 detects MAX and MIN in the window. Of course, this detection means can be shared with that in the edge creation means 102.

Reference numeral 124 denotes a subtractor, which is (MAX-MI
N) is performed. That is, it corresponds to obtaining the dynamic range in this window. Reference numeral 125 denotes a weighting unit, which is a coefficient multiplier provided to determine whether the edge information at the time of obtaining the distribution ratio (a) is to be regarded as more important or less important. This coefficient may be obtained experimentally to optimize the system, or may be determined according to the target image.

Reference numeral 126 denotes a clip unit which clips a value overflow caused by multiplication of coefficients. The distribution ratio (a) calculated in this way is output from the output terminal 127, and controls distribution of the combination of the edge information and the linear interpolation information.
By the above processing, the created edge information largely depends on the steep edge portion, and the linear interpolation information greatly depends on the flat portion.

<Modification 1 of First Embodiment> FIG.
FIG. 14 is a block diagram showing a first modification of the edge creating means 102 according to the embodiment. The modification 1 is different from the first modification in that the sorting unit 117 of the edge creating unit 102 shown in FIG. 2 is simplified. In FIG. 6, the same parts as those in FIG. In the figure, reference numeral 115 denotes an arrangement pixel number calculating means, which is a means for calculating the number of pixels for arranging MAX and the number of pixels for arranging MIN, as in the first embodiment. The feature of the first modification is that the number of pixels calculated by the arrangement pixel number calculation unit 115 is transmitted to the sorting unit 131.

The sort unit 131 determines whether to end the sort based on the number of transmitted pixels. For example, if it is assumed here that DOT _MAX is 21 pixels (the block is N = M = 8, 64 pixels), the sorting means 115 inputs a value of 21 pixels and sets each pixel of the linear interpolation information 113 to a value. It is sufficient that only 21 pixels can be limited in the descending order of the pixels, and it is not necessary to sort pixels having lower values. MAX is arranged by the dot arrangement means 116 for the upper 21 pixels, and MIN is arranged for the other 43 pixels as in the above-described embodiment. Thus, by simplifying the sorting means,
Processing can be executed at higher speed.

<Modification 2 of the first embodiment> FIG.
14 shows a second modification of the edge creating means 102 according to the embodiment.
It is a block diagram. In the second modification, the modification shown in FIG.
The difference is that the sorting means is much simpler than in Example 1.
Has become. 7, the same parts as those in FIG.
A description will be given with numbers. Reference numeral 115 denotes an arrangement pixel number calculation unit.
Then, as in the first embodiment, the number of pixels for arranging MAX is
A means for calculating the number of pixels for arranging the MIN is shown.
The feature of the second modification is calculated by the arrangement pixel number calculation unit 115.
DOT _MAX , DOT_MIN The comparator 132
And has a means to compare which pixel number is smaller
That is the point.

The DOT _MAX and DOT _MIN comparators 132 compare the number of transmitted pixels with each other to determine whether the number of pixels is smaller. The number of pixels having the smaller value is sent to the sorting means 131. Send out. In this comparison, the number of pixels may be compared with each other, or only one of the number of pixels, for example, DOT _MAX is calculated, and DOT _MAX is larger or smaller than 1/2 of the number of pixels (N × M) of the block. May be compared. For example, if it is larger than 1/2, DOT _MIN ;
OT _MAX is sent to the sorting means 131. This allows
In the sorting means 131, the order of (N × M / 2) pixels can be set at the maximum. Thus, by simplifying the sorting means, processing can be executed at higher speed.

As described above, according to the first embodiment, the pixel value in the block of the high resolution information corresponding to the target pixel of the input low resolution information is determined by the area ratio at which the target pixel value can be stored. By arranging the peripheral pixel values, this corresponds to estimating and creating edge information, and it is possible to create high-resolution information with sharp edges. In addition, since edges are created based on the sorting of the linearly interpolated pixel values, good edges can be created. Furthermore, by adaptively changing the synthesis ratio between the created edge information and the linear interpolation information, especially when the input is a natural image, it is possible to avoid a painting-like image due to artificial edge creation. be able to.

<Second Embodiment> Next, a second embodiment according to the present invention will be described.
Will be described in detail with reference to the drawings. FIG.
FIG. 9 is a block diagram illustrating a configuration of an image processing apparatus according to a second embodiment. As shown, the same parts as those in the above-described first embodiment are denoted by the same reference numerals, and description thereof is omitted. In the figure, reference numeral 201 denotes an edge creating unit according to the second embodiment, which is a _target pixel (I _xy ) of low-resolution image information (low-resolution information).
) Is created for each block of N pixels × M pixels around the center.

FIG. 9 shows the edge creating means 20 shown in FIG.
1 is a diagram showing a detailed configuration of FIG. This edge creating means 201 is a feature of the second embodiment. A portion surrounded by a broken line corresponds to the edge creating means 201. In the figure, reference numeral 210 denotes an input terminal for low resolution information, to which information indicated by 211 is input. In 211, the pixel of E corresponds to the pixel of interest, and the portion surrounded by the dashed line is the window for the neighboring pixels of the pixel of interest.

With respect to the low resolution information 211 input from the input terminal 210, the maximum value (MAX) and the minimum value (MIN) in the window are detected by the MAX / MIN detecting means 212. The detected MAX and MIN are sent to the arrangement pixel calculation means 213, and the number of pixels of each of the binary representative values for forming the edge is calculated. Assuming that the number of pixels for arranging the representative value of _MAX is DOT _MAX and the number of pixels for arranging the representative value of _MIN is DOT _MIN , the number of pixels is determined by the following equation.

DOT _MAX = (I _xy −MIN) × N × M /
(MAX-MIN) DOT _MIN = N × M-DOT _MAX On the other hand, the neighboring pixel group calculating means 214 calculates the pixel value of the neighboring pixel group of the target pixel E based on the window information. The calculation of the adjacent pixel group is a process of roughly calculating the values of the four corners of the target pixel. The values of the four corners are calculated as follows.

Upper left corner: a = αA + βB + βD Upper right corner: c = αC + βB + βF Lower left corner: g = αG + βD + βH Lower right corner: i = αI + βF + βH (α and β are coefficients) The calculated pixel group values of the four corners are: MAX (a, c, g, i) calculator
Sent to the stage 215, and the corner having the largest value among the four corners is
Selected. The selected angle is the above-mentioned DOT. _MAX Tomi
Done. Subsequently, the maximum value is calculated by the ratio calculating means 216.
Corner and the ratio of pixels adjacent to the side sandwiching the corner are calculated.
Will be issued. For example, if the angle of the pixel group having the maximum value is i
Then, F, which is a pixel adjacent to the side sandwiching the corner of i,
The ratio of the value of H is calculated, and the calculated ratio and distribution are calculated.
According to the number of pixels calculated by the placement pixel number calculating means 213,
MAX and MIN dots are arranged. 218 is out
Input terminal, and an edge inside the block as shown at 219.
The block information that created the page is output.

FIG. 10 shows the dot arrangement determined by the ratio.
FIG. Now, as shown in FIG.
Thus, the target pixel of the low-resolution information is "60"
It is assumed that the pixel value is One pixel of this pixel of interest is
It is assumed that the image is enlarged five times and five times horizontally to 25 pixels. I_xy
= 60, N = M = 5, MAX = 180, MIN = 20
The number of pixels calculated by the arrangement pixel number calculation unit 213
(DOT _MAX ) Is for six pixels. That is, as shown in FIG.
As shown in FIG.
The value of “180” is placed, and the MIN is
The value of “20” is arranged. Therefore, these six pixels are calculated as 2
In the second embodiment, it is determined which of the five pixels is to be applied.
Points.

Here, the value that takes the maximum pixel group value is the lower right corner of the pixel of interest by the above-described comparing means, and MAX dots are arranged around this corner.
The adjacent pixels on two sides sandwiching this corner are each “160”
And “80”, and the ratio of x and y shown in FIG. 10B is 2: 1. That is, while maintaining the ratio of x: y of 2: 1, the x and y are adjusted so that the number of dots becomes 6 pixels.
Is calculated. A diagonal line is drawn using the calculated x and y,
MAX (in this case, “180”) dot is arranged at a pixel located on the corner side of the oblique line. In the remaining 19 pixels in the block, MIN (in this case, “20”) dots are arranged. That is, the oblique lines correspond to the expected edge shapes.

As described above, an edge is created in the block corresponding to the target pixel. In the example of FIG.
May be triangular in the block, but may not be triangular depending on the number of dots. FIG. 11 is a diagram illustrating an example in which edges are square in a block. For example, when x> y, the area of the triangle may be less than the number of dots even when x = N (x
<Y when y = M). In this case, while maintaining the angle of the oblique line drawn when x = N, α shown in FIG.
Is moved in parallel by a distance of x (in this case, x (in this case, x) so that the area of the square shown in FIG.
= N), y, α are calculated, and as shown in (c), the pixels in the rectangle are arranged so as to be filled with MAX values, and the other pixels are arranged with MIN values.

Similarly, when the edge shape becomes a pentagon in the block, the area is MAX due to the parallel movement of the oblique lines.
The position of the oblique line is calculated so as to conform to the dot number of. By the edge creation processing described above, it is possible to create a smooth edge in the high resolution direction, that is, high resolution information, within the enlarged block. <First Modification of Second Embodiment> FIG. 12 is a block diagram showing a first modification of the edge creating means 201 according to the first embodiment. In the first modification, a plurality of edge creating means are provided, and the edge creating means is selected based on the mutual comparison of the pixel values in the window.

In FIG. 12, the same blocks as those in FIG. 9 are described with the same numbers. In the figure, reference numeral 215 denotes MAX (a, c, g, i) calculating means for calculating the maximum value of the pixel group at the four corners of the pixel of interest, as in the second embodiment. Comparison means 220
Then, based on the signal of the maximum value of the pixel group value, a comparison is made between the addition of the corner having the maximum value and the adjacent corner and the addition of two opposing corners. For example, the angle that takes the maximum value is (a,
c, g, i), if it is i, the comparison means 2
At 20, it is determined whether or not (i + c)-(g + a)> th1 (1) (i + g)-(c + a)> th1 (2) (th1 is a preset threshold). That is, in the block as shown in FIG. 9, it is roughly determined whether the edge is a diagonal edge centering on one corner or an edge extending in the vertical and horizontal directions. In the case where the above equation (1) is satisfied, since (i + c) is farther than (g + a), it can be predicted that the edge is a vertical edge.
In addition, when the expression (2) is satisfied, (i + g)
Is farther than (c + a), it is presumed to be a horizontal edge.

When it is determined that the edge is a vertical edge, in addition to the equation (1), the following equation is used: ic <th2 (3) (th2 is a preset threshold value). When it is determined that the edge is an edge, a conditional expression of ig <th2 (4) may be added to the expression (2).

The edge creation method determining means 221 shown in FIG. 12 determines whether the edge is oblique based on the above-described comparison result.
The edge creation direction is determined depending on whether the direction is vertical or horizontal. If the above condition is not met, it is determined that the edge is in the oblique direction, and the ratio calculating means 216 calculates the ratio of the pixels adjacent to the side sandwiching the maximum value corner in the same manner as in the second embodiment described above. An edge is created. If it is determined that the edge is a vertical edge or a horizontal edge, the ratio calculating means 216 determines the vertical edge as the ratio between the upper and lower pixels of the target pixel (in this case, B: H) and the horizontal edge. Is the ratio between the left and right pixels of the pixel of interest (in this case,
D: F) is obtained, and an edge is created so as to include the corner having the maximum value.

FIG. 13 is a diagram showing an example of creating a vertical edge. As shown in FIG. 13A, a pixel “100” located at the center of the window is set as a target pixel. Here, it is assumed that the resolution of the pixel of interest is vertically and horizontally 8 times. Now, it is assumed that a vertical edge is determined in conformity with the above equations (1) and (3). Therefore, based on the pixel values of “100” and “120” above and below the target pixel, x and y are set to 1: 1.2, and the square area shown in FIG. The values of x and y are calculated so as to match the number of dots.

In this example, I _xy = 100, N = M = 8,
With MAX = 180 and MIN = 40, the number of pixels (DOT _MAX ) calculated by the arrangement pixel number calculation unit 213 is 27 pixels. That is, 2 out of 64 pixels in the block
A value of “180” is arranged for seven pixels, and a value of “40” is arranged for the other 37 pixels. Since the maximum value is the lower right corner among the four corners, a diagonal line is drawn from the right based on the obtained values of x and y, and the inside of the rectangle is filled with DOT _MAX . Thus, as shown in FIG.
A vertical edge is created. The same applies to the procedure for creating edges in the horizontal direction.

<Modification 2 of Second Embodiment> FIG. 14 is a block diagram showing a modification 2 of the edge creating means 201 according to the first embodiment. The second modification provides a simpler edge creating means than the edge creating means of the second embodiment and the first modification thereof, and is effective mainly at a low magnification.

In FIG. 14, the same blocks as those in FIG. In the figure, reference numeral 230 denotes a sorting unit which sorts pixel group values set by the adjacent pixel group calculating unit 214 in descending order. Of the four corners, it suffices to be able to sort up to the second from the largest in a simplified manner. In the second modification, based on the calculated sort information, DOT is shifted from the corner having the largest value to the second largest corner.
_The feature is to arrange _MAX .

Here, the creation of an edge in the second modification will be described with reference to FIG. In FIG. 15A, the central pixel “80” is set as a target pixel. Now, when the pixel group values of the four corners are sorted in the descending order, the result is as shown in FIG. The largest is in the lower right, followed by the upper right. Here, I _xy = 80, N = M = 4, MAX = 18
With 0 and MIN = 20, the number of pixels (DOT _MAX ) calculated by the arrangement pixel number calculation means 213 is equivalent to six pixels. That is, “180” is arranged for six pixels in the block, and “20” is arranged for the other ten pixels. In this method of arranging six pixels, as shown in FIG. 15C, six pixels are arranged from the lower right to the upper right indicated by the sort information. Thereby, it is possible to easily create an edge in the block. When the largest corner and the second largest corner are diagonal among the four corners, it is determined that the edge is not a simple edge, and the edge creation processing may not be performed. .

Although several edge creating means have been described above, it is also possible to switch these means adaptively. Further, the means for creating an edge based on the state of neighboring pixels of the target pixel and the ratio of each pixel value is not limited to the method described above. Further, a method of limiting the block for performing the processing to the center of the edge portion of the image is also effective. Further, the window size and the window shape are not limited to the above-described embodiment.

As described above, according to the second embodiment, the direction, directionality, shape, and the like of the edge are estimated from the input low-resolution information, and based on the information, the sharpness of the edge is reduced. High resolution information can be created. Further, since the edge creating means is based on the magnitude relationship between the surrounding pixels and the ratio, it is possible to easily create a good edge.

<Third Embodiment> FIG. 16 is a main block diagram showing a third embodiment of the present invention. 1 will be described with the same reference numerals. In the figure, reference numeral 100 denotes an input terminal to which low-resolution image information (low-resolution information) is input. In the case of an image having a halftone, such as a natural image, unlike a character or line image created by a host computer or the like, it is considered that some LPF (low-pass filter) is applied when the image is created. Here, the resolution dependency of the input image information is considered. In a character, a line image, or the like created by a host computer or the like, an edge generated at a low resolution depends on the original resolution, and therefore, the resolution conversion includes an obstructive frequency component. Therefore, it is necessary to convert the resolution after disturbing the disturbing frequency components (after breaking the edges) and create edges (information of high-frequency components) corresponding to the new resolution.

On the other hand, a natural image is close to a resolution-free state even at a low resolution because the LPF is applied as described above. That is, since the edge is distorted from the input state, a new high-resolution edge is created in a state where the number of pixels is increased and the frequency components that interfere are small. Now, the input image information is input to the linear interpolation means 10.
1, and one pixel of the pixel of interest is interpolated into (M × N) pixels as in the above-described embodiment. Edge creation means 2
Numeral 40 denotes a means for creating an edge in an (M × N) pixel, as in the embodiment shown in FIG.
The distribution ratio to be allocated to the pixel of interest is calculated from the value and the MIN value, and an edge creation unit may be used to arrange the pixels using sorting. In an easy configuration, a threshold (for example, TH = Means for calculating (MAX + MIN) / 2) and binarizing the linear interpolation information of the (M × N) pixels with the calculated threshold value to create an edge may be used. MAX and MIN values are substituted in the block after the edge is created.

Now, this edge information is a new edge corresponding to the new resolution. In the case of a natural image, if this edge information is output as it is, it will be a pictorial image with no gradation full of edges. Then, smoothing is performed by the smoothing means 241. Smoothing can be realized by a filter.

In the above-described embodiment, a natural image is displayed by adding and synthesizing the edge information and the linear interpolation information. However, the halftone pixel value in the middle of the edge is a value generated for the LPF, that is, the original resolution. (A value that is not very reliable). Therefore, in the present embodiment that does not perform addition and synthesis with the linear interpolation information, it is possible to create an image in a state where the resolution dependency is further lost.

Therefore, according to the first, second, and third embodiments, a low-resolution image can be easily converted to a high-resolution image. A printer or a copying machine that outputs an image can be provided. The present invention may be applied to a system including a plurality of devices or to an apparatus including a single device.

It is needless to say that the present invention can be applied to a case where the present invention is achieved by supplying a program to a system or an apparatus.

[0056]

As described above, according to the present invention,
You can easily convert low resolution images to high resolution,
It is possible to provide an image processing device capable of obtaining high-quality image quality.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of an image processing apparatus according to a first embodiment.

FIG. 2 is a diagram showing a detailed configuration of an edge creating unit 102 shown in FIG.

FIG. 3 is a diagram showing a state of linear interpolation and edge creation.

FIG. 4 is a diagram showing an example of creating edge information.

FIG. 5 is a diagram showing a detailed configuration of a distribution ratio determining unit 103.

FIG. 6 is a modification example 1 of the edge creating unit 102 shown in FIG.
FIG.

FIG. 7 is a second modification of the edge creating unit 102 shown in FIG.
FIG.

FIG. 8 is a block diagram illustrating a configuration of an image processing apparatus according to the first embodiment.

FIG. 9 is a diagram showing a detailed configuration of an edge creating unit 201 shown in FIG.

FIG. 10 is a diagram illustrating an example in which dot arrangement is determined based on a ratio.

FIG. 11 is a diagram showing an example in which an edge is rectangular in a block.

FIG. 12 is a block diagram showing a first modification of the edge creating means 201 shown in FIG. 8;

FIG. 13 is a diagram illustrating an example of creating a vertical edge.

FIG. 14 is a block diagram showing a second modification of the edge creating unit 201 shown in FIG. 8;

FIG. 15 is a diagram illustrating an example of edge creation in a second modification.

FIG. 16 is a block diagram illustrating a configuration of an image processing apparatus according to a third embodiment.

FIG. 17 shows a conventional closest interpolation method.

FIG. 18 shows a conventional bilinear interpolation method.

Claims (7)

(57) [Claims] 1. A target pixel of the low resolution magnified (N × M) blocks of pixels, an image processing apparatus for converting low-resolution image to a high resolution image, the low-resolution target pixel and its of Maximum value and maximum value from surrounding pixels
Detecting means for detecting a small value ; determining means for allocating the maximum value and the minimum value detected by the detecting means in the block; and determining the allocation ratio according to the allocation ratio determined by the determining means. Most
An image processing apparatus comprising: an arrangement unit that arranges a maximum value and a minimum value . 2. The apparatus according to claim 1, wherein said determining means determines an area ratio of a maximum value and a minimum value to be arranged in the block and a number of pixels so as to preserve the density of the pixel of interest. Image processing device. 3. An interpolating means for interpolating low-resolution pixels, and a sorting means for sorting pixels interpolated by the interpolating means, wherein the arranging means according to the area ratio and the number of pixels. The maximum value and the minimum value are assigned to the pixels sorted by the sorting means.
3. The image processing apparatus according to claim 2, wherein values are arranged. 4. The arrangement means includes a maximum value and a minimum value in a block according to a magnitude relationship and a ratio between the surrounding pixels.
The image processing apparatus according to claim 1, wherein values are arranged. 5. The arrangement means according to claim 4, wherein said arranging means comprises a plurality of pixels arranged around said pixel of interest.
Among the corners, the corner having the maximum value is calculated, and the maximum value and the maximum value in the block are calculated based on the ratio between pixels adjacent to the side sandwiching the corner.
The image processing apparatus according to claim 1, wherein a minimum value is arranged. 6. The method according to claim 1, wherein the arranging means includes:
Or, depending on the ratio between the left and right pixels, the maximum value
2. The image processing apparatus according to claim 1, wherein a minimum value and a minimum value are arranged. 7. enlarged target pixel of the low resolution (M × N) blocks of pixels, an image processing apparatus for converting low-resolution image to a high resolution image, the low-resolution target pixel and its of Maximum value and maximum value from surrounding pixels
An image processing apparatus comprising: a detecting unit that detects a small value ; an arranging unit that arranges a maximum value and a minimum value in the block; and a smoothing unit that smoothes information after the arrangement.