JP3703180B2 - Image processing apparatus and image processing method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image processing apparatus that scales input image information and converts the resolution from low resolution information to high resolution information. And image processing method It is about.
[0002]
[Prior art]
Conventionally, various methods have been proposed as a method for converting the resolution of input low resolution information into high resolution information. These methods are based on the type of target image (for example, a multi-valued image having gradation information for each pixel, a binary image binarized by a pseudo halftone, and a binary value binarized by a fixed threshold value. The conversion processing method differs depending on the image, character image, and the like.
[0003]
As a conventional interpolation method, for example, as shown in FIG. 17, the closest interpolation method that arranges the same pixel values closest to the interpolation point, four points surrounding the interpolation point as shown in FIG. A bilinear interpolation method is generally used in which the pixel value E is determined by the following calculation according to the distance of four pixel values A, B, C, and D.
[0004]
E = (1-i) (1-j) A + i. (1-j) B + j. (1-i) C + ijD However, if the inter-pixel distance is 1, i from the A in the horizontal direction and j in the vertical direction. It is assumed that there is a distance (i ≦ 1, j ≦ 1).
[0005]
Further, when converting low resolution, multi-gradation image information into high resolution and binary image information, for example, an image output device such as a printer capable of only binary output corresponds to the resolution of the printer engine. When trying to output image information to be output, it is conceivable to input engine resolution information from the host computer in advance, but considering the development on the host computer, the burden of processing, the time required for transmission, etc. A configuration is also conceivable in which low-resolution information is input to generate and output high-resolution binary information.
[0006]
Currently, an error diffusion method is often used for pseudo halftone processing by binarization of a multi-valued image. However, when the engine resolution of the printer becomes higher, it is more advantageous in terms of processing time, circuit scale, etc. to use the dither method than the error diffusion method, which requires heavy processing. Further, with respect to image quality, particularly gradation, the difference between these methods is reduced as the resolution increases. However, the problem is resolution, and when dithering is used, gradation and resolution are contradictory conditions, and deterioration at the halftone character and natural image edges is visually It will stand out. Therefore, conventionally, a method of performing binarization by the dither method after edge enhancement has been proposed.
[0007]
[Problems to be solved by the invention]
However, the conventional method has the following problems.
[0008]
First, the method shown in FIG. 17 has an advantage that the configuration is simple. However, 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 block visually. Is conspicuous and the image quality is poor.
[0009]
Further, even when this method is used for a character, a line image, a CG (computer graphic) image, etc., the same pixel value continues for each block to be enlarged. As a result, the image becomes a conspicuous inferior image of jaggedness called jaggy. FIG. 19 and FIG. 20 are examples of resolution conversion that is double in both the vertical and horizontal directions, but the above-described deterioration increases as the magnification increases. Note that “200”, “10”, and the like in FIGS. 19 and 20 are pixel values.
[0010]
The method shown in FIG. 18 is a method that is generally used for enlarging a natural image. In this method, the image quality is averaged and smoothed, but the image quality is blurred in the edge portion and 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 part, important information may not be transmitted to the receiver due to blurring due to interpolation.
[0011]
FIG. 21 shows image information obtained by interpolating the input image information shown in FIG. 19 twice vertically and horizontally by the method shown in FIG. In this method, as is apparent from FIG. 21, the pixel values are not uniform not only in the vicinity of the oblique line but also in the oblique line itself, and blurring occurs.
[0012]
Consider a case where both resolution conversion and pseudo halftone processing are performed.
[0013]
When the image after resolution conversion shown in FIGS. 17 and 18 is edge-enhanced by filtering and subjected to dither processing, in the example shown in FIG. 17, jaggy becomes more noticeable. Further, in the example shown in FIG. 18, since the edge has already become dull, that is, the edge itself has a distance due to interpolation, it seems to have occurred at low resolution even if emphasis is applied. A sharp edge is not reproduced. Furthermore, since the filtering is performed after the number of pixels after resolution conversion has increased, the processing load increases.
[0014]
In other words, in multi-gradation resolution conversion, good resolution conversion without blurring and jaggies does not occur, and in multi-gradation resolution conversion + pseudo halftone processing, both gradation and resolution are compatible. There is a problem that such processing cannot be realized.
[0015]
The present invention has been made in view of the above-described problems. The object of the present invention is to avoid blurring due to interpolation, which is a problem particularly in natural images when converting input low resolution information to high resolution information. , Image processing apparatus capable of realizing conversion processing with good image quality And image processing method Is to provide.
[0016]
[Means for Solving the Problems]
In order to achieve the above object, the present invention performs a process of converting input low-resolution image information into high-resolution image information and increasing the number of pixels of the image information by (N × M) times. In the image processing device, detection means for detecting a maximum value and a minimum value of a pixel value from within a window composed of a plurality of pixels surrounding the target pixel of the low-resolution image information, and the maximum value as well as Based on the minimum value Between the maximum and minimum values A first calculation unit that calculates a value; an interpolation unit that interpolates one pixel of the low-resolution image information into (N × M) pixels; and a target pixel of the low-resolution image information. , Each interpolation value of (N × M) pixels after interpolation by the interpolation means, Middle Second calculating means for calculating a difference value with respect to the value; and determining means for determining a pixel value in the block of (N × M) pixels so that the difference value is closer to 0 and has a steeper edge; Is provided.
[0017]
Further, the present invention provides an image processing apparatus for converting input low-resolution image information into high-resolution image information and performing a process of increasing the number of pixels of the image information by (N × M) times. Smoothing means for smoothing low-resolution image information, and detection means for detecting the maximum and minimum pixel values from a window composed of a plurality of pixels surrounding the target pixel of the low-resolution image information And the maximum value as well as Based on the minimum value Between the maximum and minimum values A first calculating unit that calculates a value, an interpolating unit that interpolates one pixel of the smoothed image information into (N × M) pixels, and a target pixel of the low-resolution image information , Each interpolation value of (N × M) pixels after interpolation by the interpolation means, Middle A second calculating means for calculating a difference value with respect to a value, and an interpolation pixel having a positive difference value among interpolation pixels whose absolute value of the difference value is smaller than a predetermined threshold value set in advance; The value is determined to be larger than the maximum value, and for an interpolated pixel whose difference value is negative, the pixel value is determined to be smaller than the minimum value. Of the pixels whose absolute value of the difference value is equal to or greater than a predetermined threshold value, the maximum value is determined for an interpolation pixel having a positive difference value, and the interpolation pixel having a negative difference value is determined. Is determined to be the minimum value Determining means.
[0018]
Furthermore, the present invention provides an image processing apparatus that performs processing for converting input low-resolution image information into high-resolution image information and increasing the number of pixels of the image information by (N × M) times. Detecting means for detecting a maximum value and a minimum value of a pixel value from within a window composed of a plurality of pixels surrounding a target pixel of low-resolution image information; and the maximum value as well as Based on the minimum value Between the maximum and minimum values A first calculation unit that calculates a value; an interpolation unit that interpolates one pixel of the low-resolution image information into (N × M) pixels; and a target pixel of the low-resolution image information. , Each interpolation value of (N × M) pixels after interpolation by the interpolation means, Middle A second calculating unit that calculates a difference value from the value; a determining unit that determines a pixel value of each of the (N × M) pixels based on the difference value; and a pixel value determined by the determining unit is a dither method Binarizing means for binarizing according to the above, and when the difference value is positive, the determining means determines a value obtained by subtracting the difference value from the maximum value as a pixel value of an interpolation pixel, and the difference value is If negative, a value obtained by subtracting the difference value from the minimum value is determined as the pixel value of the interpolation pixel.
[0019]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments according to the present invention will be described below in detail with reference to the accompanying drawings.
<First Embodiment>
FIG. 1 is a block diagram showing a main part of the image processing apparatus according to the first embodiment of the present invention. Although it is efficient that the present image processing apparatus is mainly provided in an image output apparatus such as a printer, the image processing apparatus is an application software that realizes its function in an image processing apparatus other than the image output apparatus or a host computer. It can also be built in.
[0020]
An operation procedure of the image processing apparatus according to this embodiment will be described with reference to FIG. Here, an example will be described in which input image information is converted into information of the number of pixels of vertical N times and horizontal M times.
[0021]
In FIG. 1, reference numeral 100 denotes an input terminal of this apparatus, and low-resolution image information is input from this terminal. The input low-resolution information is stored and held by the line buffer 101 for several lines. Based on the image information for several lines, window processing in units of X × Y pixels centered on the pixel of interest is performed. Reference numeral 102 denotes a MAX / MIN detector that detects the maximum value and the minimum value from the image information in the window.
[0022]
Reference numeral 103 denotes a linear interpolation unit, in which pixels between original samplings are filled by bilinear interpolation processing (hereinafter referred to as linear interpolation processing) centering on the target pixel, and interpolation information of N times vertical and M times horizontal is obtained. create. Since this linear interpolation processing has been described in the example shown in FIG. 12, the description thereof is omitted here.
[0023]
An intermediate value calculation unit 104 calculates an intermediate value a from the detected MAX and MIN according to the following equation (1).
[0024]
a = (MAX + MIN) / 2 (1)
Reference numeral 105 denotes an adder, which calculates a difference value b between the linear interpolation information and the intermediate value a in an N × M pixel block (hereinafter referred to as a target pixel block) corresponding to the target pixel. Here, if the linear interpolation information in the pixel block of interest is DATA (i, j), the difference value b (i, j) of each pixel can be calculated by the following equation (2).
[0025]
b (i, j) = DATA (i, j) -a (2)
Reference numeral 106 denotes a contrast calculation unit, which calculates the contrast value c in the X × Y pixel from the detected MAX and MIN information according to the following equation (3).
[0026]
c = MAX-MIN (3)
The contrast value c and the difference value b (i, j) calculated in this way are transmitted to the LUT (look-up table) 107, and based on the contrast between the edge contrast and the difference value in the window. A predetermined value is output. Depending on the capacity of the memory used for the LUT and its price, it is also practical to reduce the number of bits by quantizing the information of the contrast value c and the difference value b (i, j) at the input stage to the LUT. Is. The value output from the LUT is added to the intermediate value a by the adder 108, transmitted to the output terminal 109, and output to a printer engine (not shown) or the like.
[0027]
Next, high-resolution edge creation according to the present embodiment will be described with reference to FIGS.
[0028]
In FIG. 2, a black circle indicates a pixel value in the window. Here, for simplification of explanation, the one-dimensional direction is shown, and adjacent pixels (Ix−1, y) and (Ix + 1, y) of the target pixel (Ix, y) are respectively expressed as MAX values and MIN values. Assume that there is. A solid line drawn from the MAX value toward the MIN value indicates linear interpolation information. A region indicated by an arrow centered on the pixel of interest (Ix, y) indicates a region of the pixel block of interest for which high resolution information is created this time.
[0029]
FIG. 3 shows a value ((MAX + MIN) / 2) of the intermediate value a calculated from the detected MAX and MIN by a broken line. Further, the difference value between the linear interpolation information and the intermediate value a is indicated by b, and the contrast between MAX and MIN is indicated by c.
[0030]
In the present embodiment, information in a high-resolution pixel block of interest is created based on the relative relationship between these b and c values, which is based on the logic described below.
[0031]
First, since the broken line a is an intermediate value between MAX and MIN as described above, it can be regarded as the center of the edge in the window. Since the interpolation is linear, in the density direction, the closer the pixel value of the interpolation information is to a, that is, the closer the difference value b is to 0, the spatial distance is closer to the center of the edge. It will be approaching. In other words, the distance in the density direction is replaced with a spatial distance.
[0032]
If a steep edge portion of a natural image or the like has a blurring due to linear interpolation, a steep edge must be created even with high resolution information. As described above, if the change in the difference value b in the density direction is replaced with the spatial distance from the edge center, the difference value b approaches zero (that is, the edge value approaches the edge center). The LUT should be created so as to have a steep edge angle.
[0033]
In addition, since the contrast information of MAX and MIN is input to the LUT, whether the low-resolution original information is an edge portion having a large contrast based on not only the relative value with the difference value b but also the absolute value of c, Alternatively, it can be easily determined whether the portion is a flat portion. If it is a flat part, it is not necessary to create an edge, and the sharpness of the high-resolution edge can be controlled depending on the size of the edge of the original information. Note that the LUT value may be calculated experimentally by optimizing the output engine.
[0034]
Here, since the LUT configuration is adopted, a non-linear edge shape can be freely created in the density direction. Further, since the output from the LUT does not have an absolute value called a pixel value, a high resolution pixel value can be created by addition to the absolute value a.
[0035]
FIG. 4 is a diagram illustrating a processing result when a high-resolution edge is created based on an output from an actual LUT. In the figure, the bold line portion is steep edge information newly created in the target pixel block.
[0036]
As described above, according to the present embodiment, one pixel of the low resolution information is interpolated, and the image information in the pixel block after interpolation corresponding to the target pixel of the low resolution information is set to a certain value. By calculating the difference value and determining the pixel value in the interpolated pixel block based on this difference, it is possible to create high-resolution information with efficient and good image quality.
[0037]
Further, the intermediate value a which is a constant value can be dynamically obtained from the MAX value and the MIN value, and the edge center can be easily specified.
[0038]
In the above embodiment, the high resolution information is created using the LUT. However, the present invention is not limited to this, and if edge creation can be realized by a simple calculation, for example, the difference value b is It may be calculated as an evaluation function. In addition, the size of the window may be variable so that the maximum value and the minimum value of the edge are included. Further, depending on the capacity of the LUT, three values of MAX, MIN, and difference value b may be input to determine the pixel value in the target pixel block.
<Second Embodiment>
The second embodiment according to the present invention will be described below.
[0039]
In this embodiment, a method that is particularly effective for high resolution processing of characters, line images, etc. developed on a host computer will be described. The present embodiment is particularly effective when the number of gradations in the window is two levels.
[0040]
As described above, characters and line images developed and created at a low resolution cause jaggy in the conventional interpolation method. Therefore, in an input image such as a character or line image, the low-resolution edge information is destroyed by a smoothing filter to make it resolution-free, and after linear interpolation to the desired number of pixels, an edge corresponding to the resolution is newly added. A method of creating a file is also proposed.
[0041]
Therefore, in this embodiment, edge emphasis is performed using the above-described replacement of the distance in the density direction and the spatial distance.
[0042]
FIG. 5 is a block diagram showing a main part of the image processing apparatus according to the present embodiment. In the figure, the same components as those in the image processing apparatus shown in FIG. 1 are denoted by the same reference numerals, and different parts will be described.
[0043]
In FIG. 5, low resolution information input from the input terminal 100 is accumulated and stored in a line buffer 101, and a window is created surrounding the pixel of interest based on the information for several lines. Then, the maximum and minimum values in this window are detected by the MAX / MIN detection unit 102, and the intermediate value a is calculated by the intermediate value calculation unit 104 in the same manner as the above equation (1).
[0044]
On the other hand, reference numeral 200 denotes a smoothing unit, which realizes smoothing by LPF (Low Bass Filter) as shown in FIG. 6 based on information in the window. Needless to say, the filter here is not limited to that shown in FIG.
[0045]
The low-resolution information after smoothing is in a state in which the edge that has been generated is broken, and the linear interpolation unit 103 interpolates one pixel of interest into N × M pixels in the broken state. Is done.
[0046]
Reference numeral 105 denotes an adder that calculates a difference value b between the linear interpolation information after smoothing and the intermediate value a in an N × M pixel block (hereinafter referred to as a target pixel block) corresponding to the target pixel. To do. As in the first embodiment, if the linear interpolation information after smoothing in this pixel block of interest is FILT (i, j), the difference value b (i, j) of each pixel is expressed by the following formula ( 4).
[0047]
b (i, j) = FILT (i, j) -a (4)
Then, the detected MAX, MIN, and difference value b (i, j) are transmitted to the pixel value determining unit 201, and the determined pixel value is transmitted to the output terminal 109 to be sent to a printer engine (not shown) or the like. Is output.
[0048]
FIG. 7 is a flowchart showing an operation procedure of the pixel value determination unit 201 according to the present embodiment. In step S1, it is determined whether or not the difference value b (i, j) is 0 or more. If the difference value is 0 or more, a positive value α set in advance in step S2 is determined. Compare with. On the other hand, if the difference value is negative, it is compared with the value -α in step S3.
[0049]
In other words, in steps S2 and S3, the difference value b (i, j) is changed from 0 to ± α. (Threshold) It is judged whether or not they are separated.
[0050]
In step S2, the difference value is Threshold If it is determined that α is greater than or equal to α, the pixel value determination unit 201 outputs the MAX value as the output value OUT (i, j) in step S4. Also, the difference value is 0 or more, and Threshold If it is less than α, the pixel value determining unit 201 adds a preset positive value β to the MAX value and outputs it in step S5. On the other hand, in step S3, the difference value is Threshold When it is determined that it is greater than or equal to −α and less than 0, a value obtained by subtracting β from the MIN value is output (step S6), and the difference value is Threshold If it is less than -α, the MIN value is output (step S7).
[0051]
Next, high-resolution edge creation in the present embodiment will be described with reference to FIGS.
[0052]
In FIG. 8, the black circles indicate the pixel values in the window after smoothing, and in the same way as in FIG. 2, are shown in a one-dimensional direction for the sake of simplicity, and are adjacent to the target pixel (Ix, y). It is assumed that pixel values before smoothing (Ix-1, y) and (Ix + 1, y) are a MAX value and a MIN value, respectively. A solid line drawn between the pixels indicates linear interpolation information after smoothing. Further, the region drawn with an arrow around the pixel of interest Ix, y is the region of the pixel block of interest for which high resolution information is to be created this time.
[0053]
In FIG. 9, the value a ((MAX + MIN) / 2) calculated from the detected MAX and MIN is indicated by a broken line, and the difference value between the linear interpolation information and a is indicated by b. FIG. 10 shows the processing result shown in FIG.
[0054]
That is, in FIG. 10, the thick line portion corresponds to the output from the pixel value determining unit 201, the output pixel value in the portion where the difference value b is near 0 is emphasized, and the other pixel values are replaced with the MAX value and the MIN value. ing. Similar to the theory shown in the first embodiment, the spatial distance from the edge portion is determined based on the difference value b, and edge enhancement of new high-resolution information is performed. It is over.
[0055]
The reason why edge enhancement is performed on the smoothed information is to delete low-resolution edges, and the created information becomes a good image with enhanced edges without jaggies. If an edge enhancement of the created high resolution information is to be realized by HPF (high pass filter), the pixel that has been increased by a factor of N × M must be filtered.
[0056]
11 and 12 are diagrams illustrating an example in which the edge creation method according to the present embodiment is used for edge extraction. In particular, as shown in FIG. 12, by extracting only the portion where the difference value is near 0 (± α range) and setting a value that protrudes from the other portions, the edge extracted image can be easily set. realizable.
[0057]
As described above, according to the present embodiment, the distance in the density direction relative to the edge center is replaced with a spatial distance, and edge enhancement is applied to the smoothed information, whereby a character, line image is obtained. Thus, it is possible to suppress the occurrence of jaggies that are a problem in resolution conversion and the like, and realize resolution conversion with good and clear image quality.
[0058]
Also in the present embodiment, an LUT may be used as in the first embodiment, and the values of α and β may be obtained experimentally.
[0059]
FIG. 13 is a block diagram illustrating a main part of an image processing apparatus according to the third embodiment. In the figure, the same components as those in the image processing apparatus shown in FIG. 1 are denoted by the same reference numerals, and different parts will be described here.
[0060]
In FIG. 13, the low resolution information input from the input terminal 100 is accumulated and stored in the line buffer 101, and a window is created surrounding the pixel of interest based on the information for several lines. Then, the maximum and minimum values in the window are detected by the MAX / MIN detection unit 102, and the intermediate value a is calculated by the intermediate value calculation unit 104 in the same manner as the above equation (1).
[0061]
On the other hand, in the image information in the window, the linear interpolation unit 103 interpolates one pixel of interest into N × M pixels. Further, the adder 105 calculates the difference value b between the linear interpolation information and the intermediate value a within the N × M pixel block (hereinafter referred to as the target pixel block) corresponding to the target pixel, using the above equation (2). Calculate with
[0062]
The detected MAX, MIN, and difference value b (i, j) are transmitted to the pixel value determining unit 300, and the pixel value determined here is transmitted to the comparator 301, and the dither as shown in FIG. It is compared with the binarized threshold output created based on the matrix. That is, here, binarization is performed based on the threshold value of the dither signal, and the comparison result is transmitted to the output terminal 109 and output to a printer engine (not shown) or the like.
[0063]
The pixel value determination unit 300 receives MAX, MIN, and the difference value b (i, j) and performs the following calculation. That is,
When b (i, j) ≧ 0
OUT (i, j) = MAX−b (i, j)
When b (i, j) <0 (5)
OUT (i, j) = MIN−b (i, j)
Hereinafter, edge processing in the present embodiment will be described.
[0064]
FIGS. 15 and 16 are diagrams for explaining edge processing in the present embodiment. In FIG. 15, ● marks indicate pixel values in the window. Also here, in order to simplify the explanation, the one-dimensional direction is shown, and the adjacent pixels (Ix-1, y) and (Ix + 1, y) of the target pixel (Ix, y) are the MAX value and the MIN value, respectively. Assume that In FIG. 15, a solid line drawn from the MAX value toward the MIN value indicates linear interpolation information. In addition, the region drawn with an arrow around the target pixel Ix, y indicates the region of the target pixel block for which high resolution information is to be created this time.
[0065]
FIG. 16 shows the value a ((MAX + MIN) / 2) calculated from the detected MAX and MIN by a broken line. The thick line portion indicates the output from the above-described pixel value determining unit 300 (actually, it is discontinuous when b (i, j) = 0, but for the sake of easy explanation, the solid line is continuously displayed. Pulling).
[0066]
As is clear from FIG. 16, the linear interpolation information and the bold line are mirror images in each of the positive part and the negative part of b, and the thick line locus is significantly different from the locus of the linear interpolation information. As the distance from the edge approaches, the contrast of the output value increases.
[0067]
The creation of the edge as described above is based on the following theory.
[0068]
First, in this embodiment, since the output value from the pixel value determining unit 300 is not the final output value, an edge portion that is advantageous for the dither method applied in the subsequent stage should be created. If the bold line portion shown in FIG. 16 is the final output, the information on the created pixel block of interest will lose continuity with the information on the neighboring pixel block to be created, and a large deterioration in image quality is expected.
[0069]
However, when using the dither method, especially when using a dither matrix with an enlargement ratio (N × M) or more, the density of the pixel of interest at the time of low resolution is maintained as much as possible, and dots are formed at the edges at the time of high resolution An edge shape in which the concentration is concentrated is preferable. That is, since the contrast increases in accordance with the distance from the edge, deterioration near the edge is difficult to be deteriorated by the dither method, and the influence of the dither method is affected as the distance from the edge increases.
[0070]
In other words, in the pixel-of-interest block to be created, both the resolution priority and the gradation priority are compatible.
[0071]
As described above, in this embodiment, after the pixel value is determined, gradation processing is performed by the dither method, so that the density change of the original information is less than that of edge enhancement by filtering after creating a high-resolution edge. In addition, both resolution and gradation can be achieved, and a good binary image can be easily realized.
[0072]
The present invention may be applied to a system composed of a plurality of devices or an apparatus composed of one device. Needless to say, the present invention can also be applied to a case where the present invention is implemented by supplying a program to a system or apparatus. In this case, the storage medium storing the program according to the present invention constitutes the present invention. Then, by reading the program from the storage medium to the system or apparatus, the system or apparatus operates in a predetermined manner.
[0073]
【The invention's effect】
As described above, according to the present invention, one pixel of the low resolution information corresponding to the target pixel of the low resolution information is interpolated, and the pixel after the interpolation Between the interpolated value and the maximum and minimum pixel values By determining the pixel value in the block of interpolated pixels so that the edge becomes steeper as the difference value is closer to 0 based on the difference value from the value, high resolution with efficient and good image quality Information can be created.
[0074]
Further, according to another invention, by replacing the distance in the density direction of the edge center relatively with a spatial distance, it is possible to suppress the occurrence of jaggies and realize a resolution conversion with good and clear image quality.
[0075]
Furthermore, according to another invention, after the pixel value is determined, the gradation process is performed by the dither method, so that the edge information created by the resolution conversion can be performed even when the resolution conversion and the pseudo halftone process are simultaneously performed. However, it is not deteriorated by the dither method, and both resolution and gradation can be achieved.
[0076]
[Brief description of the drawings]
FIG. 1 is a block diagram showing a main part of an image processing apparatus according to a first embodiment of the present invention.
FIG. 2 is a diagram for explaining high-resolution edge creation according to the first embodiment;
FIG. 3 is a diagram for explaining high-resolution edge creation according to the first embodiment;
FIG. 4 is a diagram showing a processing result when a high-resolution edge is created in the first embodiment.
FIG. 5 is a block diagram illustrating a main part of an image processing apparatus according to a second embodiment.
FIG. 6 is a diagram illustrating an example of an LPF (low bass filter) for smoothing.
FIG. 7 is a flowchart showing an operation procedure of a pixel value determination unit according to the second embodiment.
FIG. 8 is a diagram for explaining edge creation with high resolution according to the second embodiment;
FIG. 9 is a diagram for explaining high-resolution edge creation according to the second embodiment;
FIG. 10 is a diagram illustrating a processing result when a high-resolution edge is created in the second embodiment.
FIG. 11 is a diagram illustrating an example in which the edge creation method according to the second embodiment is used for edge extraction.
FIG. 12 is a diagram illustrating an example in which the edge creation method according to the second embodiment is used for edge extraction.
FIG. 13 is a block diagram illustrating a main part of an image processing apparatus according to a third embodiment.
FIG. 14 is a diagram illustrating an example of a dither matrix.
FIG. 15 is a diagram for explaining edge processing in the third embodiment;
FIG. 16 is a diagram for explaining edge processing in the third embodiment;
FIG. 17 is a diagram for explaining a conventional closest interpolation method;
FIG. 18 is a diagram for explaining a conventional bilinear interpolation method.
FIG. 19 is a diagram showing input information.
FIG. 20 is a diagram illustrating a processing result obtained by a conventional closest interpolation method.
FIG. 21 is a diagram showing a processing result in a conventional bilinear interpolation method.
[Explanation of symbols]
100 input terminals
101 Line buffer
102 MAX, MIN detector
103 Linear interpolation unit
104 Intermediate value calculator
105 adder
106 Contrast calculator
107 LUT (Lookup Table)
108 Adder
109 Output terminal
200 Smoothing part
201,300 pixel value determination unit
301 Comparator
302 dither matrix

Claims (7)

In an image processing apparatus that converts input low-resolution image information into high-resolution image information and performs a process of increasing the number of pixels of the image information by (N × M) times,
Detecting means for detecting a maximum value and a minimum value of a pixel value from within a window composed of a plurality of pixels surrounding the target pixel of the low-resolution image information;
First calculating means for calculating an intermediate value between the maximum value and the minimum value based on the maximum value and the minimum value;
Interpolation means for interpolating one pixel of the low resolution image information into (N × M) pixels;
Second calculation means for calculating a difference value between each interpolated value of the (N × M) pixels after interpolation by the interpolation means and corresponding to the target pixel of the low-resolution image information; ,
An image processing apparatus comprising: a determination unit that determines a pixel value in the block of (N × M) pixels so that the difference value is closer to 0 and has a steeper edge.   Furthermore, it comprises a third calculating means for calculating a contrast value of an edge by subtracting the minimum value from the maximum value, and determines whether the edge portion or a flat portion is based on the contrast value, and the determining means The image processing apparatus according to claim 1, wherein a pixel value in the block of (N × M) pixels is determined without creating an edge for a flat portion. The determination unit determines the pixel value by inputting the difference value and the contrast value and outputting a value for controlling the slope of the edge to be added to the intermediate value according to the input value (LUT). The image processing apparatus according to claim 2 , wherein In an image processing apparatus that converts input low-resolution image information into high-resolution image information and performs a process of increasing the number of pixels of the image information by (N × M) times,
Smoothing means for smoothing the low-resolution image information;
Detecting means for detecting a maximum value and a minimum value of a pixel value from within a window composed of a plurality of pixels surrounding the target pixel of the low-resolution image information;
First calculating means for calculating an intermediate value between the maximum value and the minimum value based on the maximum value and the minimum value;
Interpolation means for interpolating one pixel of the smoothed image information into (N × M) pixels;
Second calculation means for calculating a difference value between each interpolated value of the (N × M) pixels after interpolation by the interpolation means and corresponding to the target pixel of the low-resolution image information; ,
Among interpolation pixels whose absolute value of the difference value is smaller than a predetermined threshold value set in advance, for an interpolation pixel whose difference value is positive, the pixel value is determined to be larger than the maximum value, and the difference value is For interpolation pixels that are negative, the pixel value is determined to be smaller than the minimum value, and among the pixels whose absolute value of the difference value is equal to or greater than a predetermined threshold value, the interpolation pixel that has a positive difference value An image processing apparatus, comprising: a determining unit that determines the maximum value for an interpolation pixel and determines the minimum value for an interpolation pixel having a negative difference value. In an image processing apparatus that converts input low-resolution image information into high-resolution image information and performs a process of increasing the number of pixels of the image information by (N × M) times,
Detecting means for detecting a maximum value and a minimum value of a pixel value from within a window composed of a plurality of pixels surrounding the target pixel of the low-resolution image information;
First calculating means for calculating an intermediate value between the maximum value and the minimum value based on the maximum value and the minimum value;
Interpolation means for interpolating one pixel of the low resolution image information into (N × M) pixels;
Second calculation means for calculating a difference value between each interpolated value of the (N × M) pixels after interpolation by the interpolation means and corresponding to the target pixel of the low-resolution image information; ,
Determining means for determining a pixel value of each of the (N × M) pixels based on the difference value;
Binarizing means for binarizing the pixel value determined by the determining means by a dither method;
When the difference value is positive, the determining unit determines a value obtained by subtracting the difference value from the maximum value as a pixel value of an interpolation pixel, and when the difference value is negative, the difference value is subtracted from the minimum value. An image processing apparatus, wherein a value is determined as a pixel value of an interpolation pixel.   The image processing apparatus according to claim 1, wherein the interpolation unit performs linear interpolation. In the image processing method for converting the input low-resolution image information into high-resolution image information and performing a process of increasing the number of pixels of the image information by (N × M) times,
A detection step of detecting a maximum value and a minimum value of pixel values from within a window composed of a plurality of pixels surrounding the target pixel of the low-resolution image information;
A first calculation step of calculating an intermediate value between the maximum value and the minimum value based on the maximum value and the minimum value;
An interpolation step of interpolating one pixel of the low-resolution image information into (N × M) pixels;
A second calculation step of calculating a difference value between each interpolated value of the (N × M) pixel after interpolation in the interpolation step and the intermediate value corresponding to the target pixel of the low-resolution image information; ,
And a determination step of determining a pixel value in the block of (N × M) pixels so that the difference value is closer to 0 and has a steeper edge.

Images