JP2001155149A - Method and device for correcting image, and computer- readable recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image correction technology which is natural, whose processing time is short and which does not need much memory capacity, especially a pseudo filter technology. SOLUTION: An image acquiring part 102 acquires a processing object image. An extreme value detecting part 104 scans the image and detects an extreme value of luminance, etc. A data correcting part 108 performs correction so as to make image data near the extreme value approach the extreme value. Thereby, a pseudo filter effect such as a cross filter is realized in the vicinity of the extreme value.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to an image correction technique. In particular, the present invention relates to an image correction method for correcting image data and an image correction apparatus capable of using the method.

[0002]

2. Description of the Related Art A camera generally aims at faithfully shooting a subject. However, there are photographs that aim for artistic purposes and photographs that aim for special visual effects. As one of such special effects, a technique is known in which a cross filter is used to blur light in a high-luminance portion of an image radially or crosswise. 2. Description of the Related Art A cross filter used in a conventional general camera has a glass surface of a filter physically scratched, and diffused light is diffusely reflected by the scratch to obtain a cross-shaped or radial light spread.

Recently, software that allows a general user to edit an image relatively easily has become widespread. For such software, particularly photo retouching software, there is known a technique of detecting a high-brightness portion of an image, pasting a radial or cross-shaped pattern prepared in advance on the portion, and simulating photographing with a cross filter. Japanese Patent Application Laid-Open No. Hei 6-30373 discloses a pseudo-although good image by photographing at a relatively low exposure at the time of photographing, detecting a high-brightness portion of the obtained image, and pasting a pattern on the portion. A technique for realizing a simple cross filter effect has been disclosed.

[0004]

However, in the method of pasting the cross pattern, an unnatural impression may appear on the processed image as if it were cut and pasted. FIG. 1 shows a photographic image in which a cross pattern is pasted on a high-brightness portion of a flower at the lower right of FIG. As can be seen from this example, since a portion having the highest luminance is detected and a pattern is pasted on that portion, an artificial impression remains. On the other hand, in order to detect all points having high luminance to some extent and paste the cross pattern on all of them, the time required for detection is long, and the memory capacity required for processing is large.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and has as its object to provide an image processing apparatus capable of obtaining a more natural image after processing with a relatively short processing time and a small memory capacity. An object of the present invention is to provide a correction method and an image correction device that can use the correction method. This object is achieved by a combination of features described in the independent claims. The dependent claims define specific and useful aspects of the present invention.

[0006]

One embodiment of the present invention is an image correction device. The apparatus includes an image acquisition unit that reads an image, an extreme value detection unit that scans the image to detect an extreme value of the data value, and data that corrects the data value in the direction of the extreme value in the vicinity of the extreme value. And a correction unit. Here, the “extreme value” is a local maximum value or a local minimum value, and which one is defined by the definition of the data value. For example, when the brightness of an image is calculated from the data value, the maximum value may be considered. According to this configuration, first, an image is read, and an extreme value of a data value of the image is detected. If an extreme value is detected, the data value of the image is corrected in or near the extreme value. At that time, the data value is corrected in a direction approaching the extreme value.

[0007] The correction amount of the data value may be determined according to a difference between the data value and the extreme value. For example,
The larger the difference is, the larger the correction amount can be.

[0008] The image processing apparatus may further include a sensitivity adjustment unit. The sensitivity adjuster adjusts the detection sensitivity when detecting the extreme value from the data value. The image processing device may further include a range obtaining unit. The range obtaining unit obtains at least one of a specification regarding a range in which the extreme value is searched and a specification regarding a range in which the data value is corrected. As an example, there is a range specification regarding chromaticity or lightness obtained from the data value. The range specification is
It may be related to an area, for example, an area to be searched for an extreme value in an image, or an area to be corrected for the data value may be specified. The image processing device may further include a scanning direction acquisition unit. In that case,
The extremum detection unit searches for an extremum in a designated direction.

Another embodiment of the present invention is an image correction method.
The method includes the steps of reading an image, scanning the image for an extreme value of the data value, and correcting the data value in the direction of the extreme value near the extreme value. Further, a computer-readable recording medium that records a program for causing a computer to execute these methods may be provided. The above summary of the invention does not list all the features required for the present invention, and naturally, sub-combinations of these feature groups can also constitute the invention.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The following embodiments do not limit the invention described in the claims, and all combinations of features described in the embodiments are solutions of the invention. It is not always necessary for the means.

Here, a color thermal printer will be described as an example of a system incorporating the image processing apparatus according to the embodiment. First, an outline of color thermal paper will be described as preliminary knowledge.

The color thermosensitive paper has, for example, a thermosensitive layer of cyan, magenta, and yellow, and a transparent protective layer formed on a support in this order from the bottom. The thermal energy to be applied to each heat-sensitive layer to develop a color is different, and the energy of the lowermost cyan heat-sensitive layer is usually the highest, and the energy of the uppermost yellow heat-sensitive layer is usually the lowest. Therefore, the heat energy can be adjusted so that each heat-sensitive layer can be colored sequentially.

The colors of the yellow and magenta heat-sensitive layers obtained after heating are fixed by different ultraviolet rays. For example,
The maximum absorption wavelengths of the yellow and magenta heat-sensitive layers are about 420 nm and about 365 nm, respectively. When the heat-sensitive layers are irradiated with ultraviolet light having an energy peak at the maximum absorption wavelength corresponding to each, the color is fixed. Fixing prevents the recorded thermosensitive layer from re-coloring during the recording of the next thermosensitive layer.

When yellow, magenta, and cyan are recorded, two thermal energies, bias thermal energy and gradation thermal energy, are applied to each color.
The bias heat energy is slightly lower than the heat energy required for the heat-sensitive layer of each color to develop a color, and is applied to the color heat-sensitive paper prior to recording. The bias thermal energy is determined with reference to bias data described later. On the other hand, the gradation thermal energy is determined by the gradation level of the color to be recorded, and is applied after bias heating. The gradation heat energy is determined by image data described later.

FIG. 1 is a schematic configuration of a color thermal printer. The image processing apparatus according to the embodiment is realized mainly by cooperation of the controller 12 of the printer 10 and an image processing program stored in the system memory 14, as shown in FIG.

The printer 10 has a controller 12 for controlling the entire apparatus. The controller 12 can be composed of, for example, a general-purpose microprocessor. System memory 14
Stores a control program of the printer 10, a system program, and other system information. System memory 1
4 may be composed of a DRAM, an SRAM, or the like.
It may be constituted by a nonvolatile memory such as a LASH memory or an EEPROM (electrically erasable and programmable memory). In this embodiment, the system memory 14 is provided with an LUT (lookup table) which the controller 12 refers to appropriately for controlling the thermal head 32.

The controller 12 sends and receives signals such as data, commands, and status necessary for processing via the data bus 60 and the control bus 62 to and from various parts of the printer 10. The controller 12 receives a print request and image data to be printed from an external device (not shown) such as a PC via the communication I / F unit 52. In accordance with the print request, the controller 12 controls the thermal energy to be generated by the thermal head 32.

The external device control unit 46 includes an arbitrary storage device such as a memory card, a floppy disk, and a hard disk in which images are stored, and a PC having various I / O functions.
It controls an MCIA-compliant card or the like (hereinafter, collectively referred to as “external device”). The printer 10 may acquire image data to be printed from an external device.

The display unit 64 includes, for example, an LCD and its driver.
Displays the input user's instructions and the like. The operation unit 66 includes a menu button, a selection button, a numerical key, and the like for inputting a user instruction to the printer 10. When a touch panel is used, the display unit 64 and the operation unit 66 are integrally formed.

The image memory 16 has a yellow (Y) image memory 16a, a magenta (M) image memory 16b, and a cyan (C) image memory 16c. Each image memory 16a-
16c stores corresponding color data. At the time of recording on the color thermal paper 30, image data of each color is read line by line from the image memory 16 and written to the image line memory 18.

The bias line memory 20 holds one line of bias data. The bias data is set for each heating element, for example, in order to correct a variation in resistance values of a large number of heating elements (described later).

The strobe circuit 24 generates a strobe pulse having a predetermined duty ratio in accordance with an instruction from the controller 12, and sends the generated strobe pulse to the head drive unit 26. The selector 22 first reads one line of bias data from the bias line memory 20 and sends it to the head drive unit 26.
Subsequently, one line of image data is read from the image line memory 18 and sent to the head drive unit 26.

The head drive unit 26 generates drive data due to heat generation corresponding to bias data and image data for each line. An AND operation is performed between the drive data and the above-described strobe pulse, and a drive pulse is output.
When the drive data is “1”, the drive pulse also becomes “1”,
A transistor provided for each heating element of the thermal head 32 is driven. As a result, each heating element generates heat.

The roller pair 36 is bidirectionally rotated by a pulse motor 48, and causes the color thermal paper 30 to reciprocate in the transport path. In addition, a transport path for the color thermal paper 30 is formed by a pair of rollers (not shown). A thermal head 32 is disposed on the upper surface of the transport path, and a platen roller 44 is provided on the lower surface of the transport path opposite to the thermal head 32.

The thermal head 32 has a heating element array 32a arranged linearly in the width direction of the color thermal paper. This array includes a number of heating elements. The thermal head 32 has a first position at which the heating element array 32a is pressed against the color thermal paper 30 on the platen roller 44, and a second position at which a space is provided between them. Thermal head 32
Prints an image for each color for each line while the color thermal paper 30 is being conveyed, and when the color thermal paper 30 reciprocates three times, a color image is completed. Note that a thermal head may be provided for each color, in which case a color image is completed by one reciprocation.

A yellow fixing section 40 and a magenta fixing section 42 are provided at the end of the conveying path when viewed from the roller pair 36. The yellow fixing unit 40 and the magenta fixing unit 42 respectively have ultraviolet lamps 40a and 42a that emit ultraviolet light having the above-described energy peak. The lamp driving unit 28 is configured such that at the time of yellow recording, only the ultraviolet lamp 40a for yellow is used.
When recording magenta, an ultraviolet lamp 42 for magenta is used.
Only b is turned on.

The yellow fixing section 40 is provided with an illuminance sensor 50 for measuring the intensity of ultraviolet rays. Illuminance sensor 5
The measurement signal of 0 is sent to the controller 12, where necessary illuminance correction is performed. Even when the order of lamination of the heat-sensitive layers is different, the coloring heat energy of the intermediate layer to be recorded next may be adjusted based on the irradiation amount of the ultraviolet ray for fixing the uppermost layer.

FIG. 2 shows a schematic configuration of the image correction apparatus 100 incorporated in the printer 10. This device can be applied, for example, to digital images taken by a digital camera. In that case, a high-luminance portion of the digital image is detected, and the luminance around the high-luminance portion is approximated to the luminance of the detected high-luminance portion, whereby a pseudo cross filter effect can be generated.

The image correction apparatus 100 includes an image acquisition unit 102 that acquires image data from the data bus 60, an extreme value detection unit 104 that scans the acquired image data, and detects an extreme value of the data value. A data correction unit 108 for correcting the image data value in the direction of the extreme value in the vicinity of the detected extreme value, a sensitivity adjusting unit 106 for adjusting the detection sensitivity when detecting the extreme value, and detecting or detecting the extreme value. Range specification unit 11 for reflecting the user's range specification in data correction
0 and a scanning direction acquisition unit 112 for acquiring a direction in which image data is scanned when an extreme value is detected. Sensitivity adjuster 10
6. The range designation unit 110 and the scanning direction acquisition unit 112 both acquire the user's instruction via the operation unit 66. The corrected image data is output from the data correction unit 108 to the data bus 60 again. Each component of the image correction apparatus 110 may be configured by a software module, may be configured by a hardware device, or may be configured by an arbitrary combination thereof. Further, the scanning of the image may be performed by the image acquisition unit 102 instead of the extreme value detection unit 104, and the adjustment of the detection sensitivity may be integrally performed by the extreme value detection unit 104 instead of the sensitivity adjustment unit 106. The same applies to the correction of data, and this processing may be performed in the extreme value detection unit 104. The method of such a configuration is
For example, it depends on software design, and the degree of freedom in design is easily understood by those skilled in the art.

FIG. 4 is a flowchart showing the procedure of data correction by the image correction device 110. If the outline of the correspondence relationship between each component in FIG. 3 and the processing in FIG. 4 is shown, the image acquisition unit 102 performs the processing S10 in FIG. Extreme value detection unit 104
Correspond to the processes S12, S14, S18, S24, S2 in FIG.
8 and S30. The sensitivity adjustment unit 106 performs the processing S16 and S26 of FIG. Both the range designation unit 110 and the scanning direction acquisition unit 112 act on the extreme value detection unit 104,
This affects the processes S28 and S30 in FIG.

As shown in FIG. 4, first, the image processing unit 102 acquires image data to be processed from the data bus 60 (S
10). The extreme value detection unit 104 initializes a parameter max related to the detection of the extreme value and the correction of the data to 0 (S12). Next, the extreme value detection unit 104 reads the luminance Y of the first pixel of the acquired image data (S14).

FIG. 5 shows a scanning direction 122 for detecting extrema in an image 120. In this example, two scans are performed in the direction of the x-axis. The directions of these two scans differ by 180 °.

FIG. 6 shows the special effects of the image to be obtained in the scanning direction of FIG. Here, the correction is performed so that the high luminance portion 130 has two light bands 132 in the scanning direction in FIG. In other words, this light band 132 is the light bleeding by the pseudo filter, and the effect of the pseudo cross filter can be obtained by scanning in four directions as described later.

FIG. 7 shows the luminance Y of the first few pixels of the image data. Here, a0 to a9 are listed as pixel addresses. It is assumed that the luminance Y4 of the pixel at the address a4 is higher than the luminance Y0 of the pixels at all other addresses. By the process S14 in FIG. 4, the extreme value detection unit 104 reads the luminance Y0 of the pixel at the address a0, which is the first pixel.

Subsequently, the sensitivity adjusting unit 106 multiplies the parameter max by the sensitivity parameter α, and sets this as a new parameter max (S16). For the first pixel a0, since the parameter max is set to 0 in S12, the new parameter max obtained in S16 is also 0. Here, α is assumed to be α = ２.

Next, the extreme value detecting unit 104 compares the current luminance Y of the pixel with the value of the parameter max (S1).
8). Now, since Y> max, the process proceeds to S24, and the luminance Y0 of the pixel at the address a0 is stored as a new value of the parameter max. Subsequently, the sensitivity adjustment unit 10
6, the sensitivity parameter α is adjusted. Next, the sensitivity parameter α is adjusted by the sensitivity adjustment unit 106 (S26). However, here, α is kept at 2/3,
An example of changing α will be described later.

Subsequently, the extremum detector 104 determines whether the pixel just processed is the last pixel to be processed (S30). Since there are still pixels to process,
The process proceeds to S28. Here, the address of the pixel is incremented, and the process returns to S14, where the address a of the next pixel is
The luminance Y of 1 is read. In S16, since the current parameter max is Y0, a new parameter ma is set.
x becomes Y0 × 2/3. On the other hand, the luminance Y of the current pixel is Y
Since it is 0, the process branches to S24 again in S18, and returns to S14 through the processes of S26, S30, and S28.

The same processing is continued up to the address a4. In the processing S24 relating to the address a4, the parameter m
The value of ax is Y4. After the subsequent reading of the pixel at the address a5, the process of S18 branches to S20 for the first time. In S20, the data correction unit 108 corrects the current luminance Y5 of the pixel a5. Here, Y5 + β (m
ax-Y5) is set as a new Y5. That is, a correction amount according to the difference between the parameter max regarding the extremum and the current luminance Y is added to the current luminance. β can take various values, but here, for example, is set to ２. Thereafter, the data correction unit 108 updates the value of Y5 to a new value of Y5 (S22).

FIG. 8 shows a parameter m indicated by a dashed line.
The relationship between the locus 140 of ax and data correction is shown. The luminance Y5 (= Y0) of the pixel at the address a5 has been raised to Y'5 by the processing in S20. Similar processing is performed for the pixels at addresses a6 and a7, and the luminance is increased to Y'6 and Y'7, respectively. The value of the parameter max is reduced to 2/3 by the processing of S16 every time the address of the pixel is incremented. In the example of FIG. 8, the value of the new parameter max obtained by the processing of S16 falls below Y0, and as a result, the processing for the pixels at addresses a8 and a9 is performed at addresses a0 and a1.
And so on. Thus, the processing for one scanning direction is completed.

FIG. 9 shows the luminance of each pixel when data correction in one scanning direction is completed. As shown in the figure, since the luminance Y of the pixel at the address a4 takes a maximum value,
The luminance Y of the pixels at the subsequent addresses a5, a6, and a7 is increased. Since the value of the parameter max is larger as the address is closer to the address a4, a larger data is corrected for a pixel closer to the address a4, that is, the address a5.

Subsequently, the processing of FIG. 4 is performed for the reverse scanning direction. FIG. 10 shows the relationship between the locus 140 of the parameter max and the data correction when the scanning direction is reversed. This time, the processing of FIG. 4 is performed sequentially from the address a9 to the address a0. Address a9 to address a4
, The luminance Y of the pixel at each address is greater than the new parameter max obtained by the processing of S16, so the locus 14 of the parameter max
0 is a line connecting the luminance Y of the current pixel.

On the other hand, when the processing proceeds to the address a3, the value of the parameter max is changed to the luminance Y3 (= Y
0). Therefore, the luminance Y3 of the address a3 is raised to Y'3 by the processing of S20.
This correction amount is the same as the correction amount given to the pixel at address a5 in FIG. Similarly, address a
2, the same correction amount as the address a6 is given, and the pixel at the address a1 is given the same correction amount as the address a7. Since the value of the parameter max is settled at Y0 at the address a0, the data correction for the address a0 is not performed, and a series of correction processing is completed.

FIG. 11 shows the luminance Y of each pixel obtained as a result of completing the data correction in the two scanning directions.
As shown in the figure, it can be seen that the pixel value at the address a4, that is, the luminance of the neighboring pixels centered on the maximum value of the luminance is corrected in the direction of the maximum value.

FIGS. 12A, 12B and 12C show pseudo filter effects obtained for different scanning directions 150, 152 and 154, respectively. In FIG. 12A, scanning is performed in two directions from the top to the bottom and from the bottom to the top, and as a result, light bands 132 of the high-luminance portion 130 of the image are formed in the two directions. In FIG. 12B, one scan in the right direction causes the light band 13
2 is formed only in the right direction. In FIG. 12C,
Scanning is performed in four oblique directions, and a light band 132 of the high-luminance part 130 extends in four directions according to the scanning direction. As described above, by setting the scanning direction according to the direction in which the user wants to extend the light band 132, it is easy to obtain a desired special effect. According to this embodiment, a pseudo filter effect can be obtained in conjunction with the very simple process of scanning an image, which is extremely advantageous in both processing speed and memory capacity required for processing. Also, since smooth data correction is realized centering on the maximum value of luminance originally included in the image, extremely natural data correction is realized.

In the above processing, the user can specify data correction in more detail via the range specifying section 110. For example, by designating a partial area of an image, data can be corrected by scanning only pixels included in that area. In this case, for example, by designating only the flower garden portion in the image of FIG. 1, it is possible not to perform the filtering process on the light radiated from between the trees, for example. As another example, it is conceivable that scanning of an image is performed on the entire image, but only an image area where data is to be actually corrected is specified. In that case, the user can confirm the effect of the data correction while designating various regions.

As another example of the range designating section 110, there is designating a range of brightness of pixels to be corrected. For example, even if the luminance of a certain pixel has a local maximum value in the vicinity thereof, if the value itself is extremely small, the correction of data for the pixels in the vicinity may be skipped. In that case, the filter effect is not excessively applied to the pixels whose luminance is not so high, and the sharpness of the entire image can be maintained. As yet another example, a range can be specified for chromaticity.

FIG. 13 shows a rule for adjusting the sensitivity parameter α. Here, it is assumed that the parameter max is 8-bit data and takes a value of 0 to 255. The sensitivity parameter α has a constant value of 160/256 when the parameter max is in the range of 0 to 128, and the parameter max is 128.
160/256 to 250/256 when at ~ 240
And the parameter max is 240 to 2
When it is at 55, it is assumed that a constant value of 250/256 is taken. Therefore, the value of the sensitivity parameter α increases as the luminance of the pixel currently being processed increases (S16), and the data correction amount can be increased (S20).

FIG. 14 shows a locus 140 of the parameter max obtained when the sensitivity parameter α is adjusted according to FIG. 13 in the process of S26 in FIG. Here, it is assumed that the luminance Y4 of the pixel at the address a4 is a sufficiently large value. In this case, since the sensitivity parameter α is relatively close to “1” near the address a4, the parameter max is
Is small. However, as the value of the parameter max decreases, the amount of decrease of the parameter max gradually increases due to the processing of S26 and S16. In FIG. 14, the value of the parameter max is Y at addresses a0 and a8.
The example which returns to 0 is shown.

FIG. 15 shows the luminance of each pixel corrected according to the value of the parameter max in FIG. As shown in the figure, a pixel closer to the address a4 is given a larger correction amount than the processing in FIG. 11, and a large pseudo filter effect can be obtained as a whole. The merit of this method is that a large amount of data correction can be taken in the vicinity of a pixel having a larger maximum value, and a small amount of data correction can be taken in the vicinity of a pixel having a not-so-large extreme value. That is, a large filter effect, that is, a longer band of light can be given to a portion where the brightness is larger in the image. As a result, a more special effect is realized.

FIG. 16 is a flowchart showing the procedure of data correction taking into account hue. In this figure, the same processes as those in FIG. The new process in FIG. 16 is the determination of the hue by the extreme value detection unit 104 (S40). This process focuses on the local approximation of the image, and is a very simple procedure that produces a high effect. That is, even if the luminance of the pixel currently being processed is larger than the parameter max in S18 (N in S18), if the hue is not the target hue (N in S40), the processing is skipped and the next address is skipped. To move to the point. In this case, the parameter m
Since the value of ax remains relatively small without being updated, the luminance Y is set to the parameter ma in S18 for the next pixel.
x (Y in S18) is reduced, and as a result, data correction (S20) is less likely to occur. Meanwhile S4
When it is determined that the pixel being processed has the target hue at 0 (Y in S40), the value of the parameter max is increased (S24). Therefore, for the next pixel, S
The probability of branching from S18 to S20 is increased, and data correction is more likely to occur. When an image is locally examined, generally, data values of adjacent pixels generally have a high degree of similarity. Therefore, the processing of FIG. 16 realizes a filtering effect centering on pixels having higher luminance and entering a target hue. be able to.

Now, description will be made assuming that the target hue is approximately green. Therefore, first, the hue angle display is defined. Now, the image data is 8-bit luminance and color difference Y, C respectively.
It is assumed that they are represented by r and Cb. These parameters and RGB conversion equations are generally

Y = 0.2990R + 0.5870G + 0.1140B Cr = 0.5000R−0.4187G−0.0813B + 128 Cb = −0.1687R−0.3313G−0.5000B + 128 Here, “128” of Cr and Cb indicates that the data is 8
It is added for convenience represented by bits 0 to 255.
Therefore, cr = Cr-128 cb = Cb-128 can be taken as an example of the hue coordinate axes used in the present embodiment, and the hue angle indication φ can be defined as φ = tan ⁻¹ (cr / cb). On the other hand, the luminance rr can be described as rr = (cr ² + cb ² ) ^1/2 . Here, for example, 190 ° <φ <260
When ° and rr> 15, it can be determined that the target hue is obtained. According to this method, it is possible to add a pseudo filter effect only to pixels of a specific color,
It can satisfy various demands of users for special effects.

Although the embodiments have been described above, the technical scope of the present invention is not limited to these descriptions. It is understood by those skilled in the art that various changes or improvements can be made to these embodiments.

As a first modification, the embodiment focuses on the luminance of pixels. However, this may be the chromaticity, brightness, or other various attributes of the pixel. By selecting various attributes, various special effects can be obtained. Further, the unit for data correction does not need to be a pixel, and may be, for example, an image area or a block having a certain extent.

As a second modification, a table of hue angle display φ, Cr, and Cb may be created in advance for hue determination to shorten the calculation time.

As a third modification, the value of the parameter β used at the time of data correction (S20) in FIG. 4 may be variable. Also in this case, any setting can be made, such as increasing the value of the parameter β as the difference between the parameter max and the current luminance Y increases.

As a fourth modification, the embodiment has been considered for application to a color thermal printer. However, this may naturally be another type of printer, an image-related device such as a digital camera other than a printer, or a wide variety of devices. It may be a general computer. For example, the present invention can also be realized by implementing a program for performing the processing of the embodiment on a personal computer. This program can be provided on a storage medium such as a floppy disk. When applied to a digital camera, the data correction of the present embodiment may be applied to an image read from a memory card or the like into the digital camera, in addition to an image captured by the digital camera.

[0058]

FIG. 17 shows the result of applying the data correction according to the present embodiment to the original image of the image shown in FIG. As shown in the figure, a very natural and effective image was obtained.

[0059]

According to the present invention, a natural and effective image correction technique can be realized with a relatively short processing time and a relatively small memory amount.

[Brief description of the drawings]

FIG. 1 is a photograph showing a halftone image in which an image corrected by a conventional data correction technique is displayed on a display.

FIG. 2 is a configuration diagram of a color thermal printer according to the embodiment.

FIG. 3 is a configuration diagram of an image correction device according to an embodiment.

FIG. 4 is a flowchart illustrating a data correction processing procedure performed by the image correction apparatus according to the embodiment;

FIG. 5 is a diagram showing a scanning direction for searching for an extremum in the embodiment.

FIG. 6 is a diagram illustrating an example in which a pseudo filtering effect is applied to a high-luminance portion in the horizontal direction.

FIG. 7 is a diagram illustrating luminance Y of pixels at addresses a0 to a9.

FIG. 8 is a diagram illustrating data correction performed when pixels are scanned from left to right.

FIG. 9 is a diagram illustrating luminance of a pixel obtained by data correction performed as a result of scanning the pixel from left to right.

FIG. 10 is a diagram illustrating data correction performed when a pixel is scanned from right to left.

FIG. 11 is a diagram showing corrected pixel luminance finally obtained as a result of scanning a pixel from two directions.

FIG. 12 is a diagram illustrating a pseudo filter effect to be obtained corresponding to various scanning directions.

FIG. 13 is a diagram showing a relationship between a parameter max relating to an extreme value and a sensitivity parameter α.

FIG. 14 shows a parameter m obtained based on the relationship shown in FIG.
FIG. 4 is a diagram illustrating a relationship between ax and luminance of a pixel.

FIG. 15 is a diagram illustrating luminance of a pixel after correction finally obtained with respect to a parameter max in FIG. 14;

FIG. 16 is a flowchart illustrating a procedure for correcting data of only a pixel of a target hue.

FIG. 17 is a photograph showing a halftone image in which an image obtained by the data correction according to the embodiment is displayed on a display.

[Explanation of symbols]

REFERENCE SIGNS LIST 10 printer 12 controller 14 system memory 16 image memory 64 display unit 66 operation unit 100 image correction device 102 image acquisition unit 104 extreme value detection unit 106 sensitivity adjustment unit 108 data correction unit 110 range designation unit 112 scanning direction acquisition unit 120 image 122 scanning Direction 130 High-intensity part 132 Band of light 140 Locus of parameter max

 ──────────────────────────────────────────────────続 き Continued on the front page F-term (reference) 5B057 BA24 CA01 CA16 CB01 CB16 CE11 CE20 DC22 5C023 AA08 AA31 AA38 AA40 BA03 CA02 CA09 EA03 EA05 5C066 AA05 BA20 CA21 DA01 DC01 EA07 EA11 ED12 GA04 GA05 GB07 HA04 MP03 KD07 PP52

Claims (10)

[Claims] An image acquisition unit that reads an image; an extreme value detection unit that scans the image to detect an extreme value of a data value; and corrects the data value in the vicinity of the extreme value in the direction of the extreme value. An image correction device, comprising: 2. The image correction apparatus according to claim 1, wherein the extreme value is a maximum value of brightness obtained from the data value. 3. The image correction apparatus according to claim 1, wherein a correction amount of the data value is determined according to a difference between the data value and the extreme value. 4. The image correction apparatus according to claim 1, further comprising a sensitivity adjustment unit that adjusts the detection sensitivity when detecting the extremum from the data value. 5. The apparatus according to claim 1, further comprising a range obtaining unit configured to obtain at least one of a range for searching for the extremum and a range for correcting the data value.
The image correction device according to any one of the above. 6. The image correction apparatus according to claim 5, wherein the range obtaining unit obtains a range specification regarding chromaticity obtained from the data value. 7. The image correction apparatus according to claim 5, wherein the range obtaining unit obtains a range specification related to lightness obtained from the data value. 8. The apparatus according to claim 1, further comprising a scanning direction acquisition unit for acquiring an instruction of the scanning direction, wherein the extreme value detection unit searches for the extreme value in the designated direction. 8. The image correction device according to any one of 7. 9. A step of reading an image, scanning the image to search for an extreme value of the data value, and correcting the data value in the direction of the extreme value in the vicinity of the extreme value. An image correction method comprising: 10. A computer-readable recording medium, comprising: reading an image; scanning the image to find an extreme value of the data value; and reading the data value in the vicinity of the extreme value. A recording medium characterized by recording a program for causing a computer to execute the following steps: