JP5426953B2 - Image processing apparatus and method - Google Patents

Description

  The present invention relates to image processing for correcting image quality degradation caused by an image output apparatus.

  Opportunities to output an image represented by digital image data using an image output device such as a printer have increased. In general, the frequency characteristics of an image are modulated by the frequency characteristics of the image output device, and the image quality of the output image is degraded. The frequency characteristic of the image output device is represented by MTF (modulation transfer function) which is an absolute value of an optical transfer function (OTF).

  A restoration process is known in which image quality degradation caused by MTF is restored by image processing. In the restoration process, the image signal is generally corrected by filtering in real space by convolution integration. For example, Patent Document 1 is an invention that reduces image quality deterioration of an output image in a high-frequency region by controlling the level of correction based on a difference in level between a target pixel and the pixels before and after the target pixel and correcting the target pixel. Is disclosed.

  In addition, it is known that image quality deterioration due to MTF is not uniform in an image. Patent Document 2 discloses an invention in which the degree of correction (hereinafter referred to as MTF correction) is changed in accordance with the direction when image quality degradation caused by MTF varies depending on the direction, that is, when the MTF has anisotropy. Patent Document 3 discloses an invention in which the degree of MTF correction is dynamically changed in accordance with the pixel position when image quality degradation caused by MTF varies depending on the position.

  There is also known a method of changing the filter processing depending on the type of image and adjusting the degree of correction. In the invention of Patent Document 4, sharpness enhancement processing is performed on characters and line drawings in order to improve the quality of characters and line drawings, and smoothing processing is performed on gradation images such as photographs in order to improve gradation characteristics.

  However, the degree of image quality degradation due to MTF may vary depending on the feature quantity such as the amplitude value and average luminance value for each frequency component of each pixel. That is, even for pixels at the same position output using the same device, the MTF changes due to differences in luminance and the like, and the degree of image quality deterioration differs.

  A difference in image quality degradation caused by MTF due to a difference in luminance will be described with reference to FIGS. FIG. 1 shows two sinusoidal luminance changes with equal average luminance values and different luminance amplitudes. FIG. 2 shows two sine wave-like luminance changes having the same luminance amplitude and different average luminance values. An image output apparatus such as a printer may have a different MTF when the amplitude of the input data as shown in FIG. 1 is different. In the example of FIG. 1, the deterioration when the amplitude is large is small. Also, as shown in FIG. 2, the MTF may be different if the average luminance value of the input data is different. In the example of FIG. 2, the deterioration is large when the average luminance value is large.

  It will be described with reference to FIG. 3 that the MTF varies depending on the feature quantity such as the luminance amplitude and the average value. As shown in FIG. 3, the MTF differs depending on the feature quantity such as the luminance amplitude and the average value, and the degree of image quality deterioration also changes. FIG. 3 (a) shows an example of the MTF when the average luminance values are the same and the amplitudes are different, and FIG. 3 (b) shows an example of the MTF when the amplitudes are the same and the average luminance values are different.

  As described above, when the MTF changes depending on the feature amount, there is a problem that the conventional MTF correction cannot perform the MTF correction according to the feature amount. The following can be cited as an example in which the degree of image quality degradation caused by MTF changes depending on the feature amount. In an ink jet printer, the amount of dot shot in the shadow portion changes due to a change in luminance, and the influence amount of the dot gain on the highlight portion changes to cause a change in density in the highlight portion, thereby changing the MTF.

  The method of controlling the strength of the degree of MTF correction based on the difference between the level of the pixel of interest and the pixels before and after it in Patent Document 1 is based on the MTF of the image output device in order to reduce image quality degradation in the high frequency range. It is impossible to accurately correct image quality degradation caused by. Also, the method of changing the filter processing depending on the type of image cannot provide an appropriate degree of MTF correction when the original is to be faithfully reproduced.

  In general, the brightness of a fine portion in the high frequency range of the image output from the printer is lower than that of the original image. That is, the luminance changes between the original image and the output image, which contributes to image quality degradation. The factor of the luminance change varies depending on the printer.

  For example, in an ink jet printer, after ink has landed on a recording medium, ink bleeding occurs and dots are spread unnecessarily. Further, light is scattered in the recording medium, and unnecessary spread occurs in the reflected light from the recording medium. The former is called mechanical dot gain, and the latter is called optical dot gain. The luminance of the image changes due to these factors.

  In digital photo printers, the photosensitive material is exposed with a light beam to record an image. However, due to optical errors, the spot of the light beam on the photosensitive material surface is slightly larger than the appropriate spot size, and the image is recorded. Changes in brightness.

The following luminance correction is known for such a luminance change in the output image of the image output apparatus.
A method of calculating the correction amount and adding the correction amount to the input pixel value;
A method of converting image data before output by a lookup table (LUT) created from the difference between the image data before output and the image data read from the output image,
A method of correcting an image signal by filtering in real space.

  The invention of Patent Document 5 outputs a density pattern of a gradation patch of a solid image, and creates a correction table from the density pattern reading result and correction data representing a difference between predetermined characteristics. Then, the correction value of the correction table corresponding to the input value is added to the input value to perform density correction. Accordingly, it is assumed that density correction can be performed even when an image based on image data acquired from other than the image reading apparatus is formed.

  The invention of Patent Document 6 calculates the correction amount using the average value of image data, the signal values of highlight points, shadow points, and the like as feature amounts. Then, density correction is performed by adding a correction amount to the input value. At this time, it is considered that the main subject included in the image includes a region where the change in the image data is large, and the region where the change in the image data is large often includes important information which is observed by the observer of the image. That is, density correction is performed by using, as one of the feature amounts, a high-frequency component that increases in a region where the change in image data is large, or an average value of image data weighted according to a local dispersion value.

  However, the degree of change in luminance of the output image may change depending on a feature quantity such as an amplitude value or an average luminance value for each frequency component of the pixel.

  An example of a sine wave chart will be described with reference to FIG. The three sine wave charts shown in FIG. 11 have predetermined amplitude values and average luminance values, and have different frequencies. When such a sine wave chart is output by the image output device and the average luminance value of the output sine wave chart is plotted, a curve as shown in FIG. 12 is obtained. That is, the average luminance value of the output image decreases as the frequency component becomes higher.

  An example of a sine wave chart will be described with reference to FIG. The two sine wave charts shown in FIG. 13 have a predetermined average luminance value and frequency, and have different amplitude values. When such a sine wave chart is output by the image output device and the average luminance value of the output sine wave chart is plotted, a curve as shown in FIG. 14 is obtained. An example of a sine wave chart will be described with reference to FIG. The two sine wave charts shown in FIG. 15 have predetermined amplitude values and frequencies, and have different average luminance values. When such a sine wave chart is output by the image output device and the average luminance value of the output sine wave chart is plotted, a curve as shown in FIG. 16 is obtained. As shown in FIGS. 14 and 16, the decrease in the average luminance value of the output image changes according to the amplitude value and the average luminance value for each frequency component.

  As described above, the luminance change of the output image depends on the characteristic amount such as the frequency characteristics of the image data, the amplitude value for each frequency component, and the average luminance value. In other words, the degree of image quality deterioration changes according to the feature amount. The luminance change depending on the feature amount cannot be corrected by a general luminance correction method. That is, the invention of Patent Document 5 can correct based on the density change of the image output device, but cannot correct the density change that differs for each frequency range. Further, although the invention of Patent Document 6 considers the frequency component of the image data, it does not consider the density change of the image output device, and of course does not consider the correction according to the amplitude value and the average luminance value for each frequency component.

JP 05-075859 A Japanese Patent Laid-Open No. 11-308460 Japanese Patent Laid-Open No. 2002-190017 JP2005-027016 JP 2001-251512 JP JP2001-257883

  It is an object of the present invention to suppress image quality degradation caused by optical transfer characteristics that change depending on the feature amount of an image for each pixel.

  Another object of the present invention is to suppress image quality degradation caused by a change in luminance of an output image depending on a feature amount of an image for each pixel.

  The present invention has the following configuration as one means for achieving the above object.

The image processing according to the present invention calculates the image feature amount for each frequency component from the image data obtained by decomposing the input image data into frequency components, using a pixel value corresponding to the pixel of interest . An optical transfer function corresponding to the feature amount is acquired for each frequency component from a storage unit that stores an optical transfer function for each frequency component, and the input image is corrected from the optical transfer function acquired for each frequency component. A correction filter for the frequency space to be generated, converting the correction filter for the frequency space into a correction filter for the real space, performing correction for each pixel of the input image using the correction filter for the real space, and after the correction This image is output to the image output device.

Further, the image feature amount for each frequency component is calculated from the image data obtained by decomposing the input image data into frequency components, using the pixel value corresponding to the pixel of interest, and the luminance change amount of the output image by the image output device For each frequency component, a luminance change amount corresponding to the feature amount is acquired for each frequency component, and a luminance change amount is selected as a luminance correction coefficient from the acquired luminance change amount, A luminance correction coefficient is used to correct the luminance for each pixel of the input image, and the corrected image is output to the image output device.

  According to the present invention, it is possible to suppress image quality degradation caused by optical transfer characteristics that change depending on the feature amount of an image for each pixel.

  Further, according to the present invention, it is possible to suppress image quality degradation caused by a change in luminance of the output image depending on the feature amount of the image for each pixel.

The figure explaining the difference in the image quality degradation resulting from MTF by the difference in brightness. The figure explaining the difference in the image quality degradation resulting from MTF by the difference in brightness. The figure explaining that MTF changes with feature-values, such as a luminance amplitude and an average value. FIG. 2 is a block diagram for explaining a configuration example of an image processing apparatus according to the first embodiment. The flowchart explaining MTF correction. The flowchart explaining the detail of the process of a feature-value calculation part. FIG. 3 is a block diagram for explaining an example configuration of an image processing apparatus according to a second embodiment. The flowchart explaining calculation of MTF data. The figure which shows an example of the chart for MTF measurement. The figure which shows an example of the chart for MTF measurement. The figure which shows an example of a sine wave chart. FIG. 12 is a diagram showing a curve of luminance values obtained by output and measurement of the sine wave chart shown in FIG. The figure which shows an example of a sine wave chart. FIG. 14 is a diagram showing a curve of luminance values obtained by measurement and output of the sine wave chart shown in FIG. The figure which shows an example of a sine wave chart. FIG. 15 is a diagram showing a curve of luminance values obtained by output and measurement of the sine wave chart shown in FIG. FIG. 6 is a block diagram for explaining an example configuration of an image processing apparatus according to a third embodiment. The flowchart explaining brightness correction. The figure explaining an example of the brightness | luminance change amount which a brightness | luminance change amount memory | storage part memorize | stores. 10 is a flowchart for explaining details of luminance correction (S1208) of a target pixel.

  Hereinafter, image processing according to an embodiment of the present invention will be described in detail with reference to the drawings.

[Device configuration]
A configuration example of the image processing apparatus according to the first embodiment will be described with reference to the block diagram of FIG.

  Connected to the image processing apparatus 100 are an image output apparatus 103 such as an ink jet printer or an electrophotographic printer, and an image storage unit 102 for storing images. The image storage unit 102 may be a server device connected via a network. The image input unit 101 inputs image data of an image to be output by the image output device 103 (hereinafter referred to as an original image) from the image storage unit 102.

  The feature amount calculation unit 104 has the following configuration to calculate the feature amount of the target pixel of the original image. The frequency conversion unit 105 decomposes the image data into frequency components using wavelet conversion or the like. The amplitude acquisition unit 106 acquires an amplitude value as a feature amount for each frequency component from the image data decomposed into frequency components using a pixel value corresponding to the target pixel. The average luminance acquisition unit 107 calculates an average luminance value as a feature amount from pixel values in a predetermined area near the target pixel including the target pixel.

  The MTF storage unit 108 stores data indicating the MTF that is the absolute value of the optical transfer function (OTF) of the image output device 103 (hereinafter referred to as MTF data). The MTF acquisition unit 109 receives the amplitude parameter and the average luminance parameter that are the feature amount of the target pixel, and acquires MTF data corresponding to the feature amount of the target pixel from the MTF storage unit 108. The correction filter creation unit 110 creates a frequency space correction filter using the MTF correction filter coefficient calculated from the MTF data. The correction filter conversion unit 111 converts the frequency space correction filter into a real space correction filter.

  The image conversion unit 112 performs a real space correction filter and image conversion (hereinafter referred to as MTF correction) for convolution integration of the image data of the original image. The image output unit 113 outputs the MTF-corrected image data to the image output device 103 to print the image.

[Image processing]
The MTF correction will be described with reference to the flowchart of FIG.

  The image input unit 101 inputs image data of the original image (S201). The feature amount calculation unit 104 selects the MTF correction target pixel (target pixel) from the image data of the input image (S202), and calculates the feature amount of the target pixel (details will be described later) (S203). The MTF acquisition unit 109 acquires, from the MTF storage unit 108, MTF data for each frequency component of the image output device 103 according to the calculated feature amount (amplitude value for each frequency component and average luminance value) (S204).

  Next, the correction filter creation unit 110 calculates a filter coefficient for MTF correction from the acquired MTF data for each frequency component, and creates a correction filter for the frequency space (S205). The filter coefficient for MTF correction is an image before the image output device 103 outputs in order to prevent deterioration of the frequency characteristics of the output image due to the MTF of the image output device 103 (match the frequency characteristics of the output image and the original image). Is a coefficient to compensate for the frequency characteristics.

  Next, the correction filter conversion unit 111 converts the correction filter in the frequency space into the real space by inverse Fourier transform or the like to generate a real space correction filter (S206). The image conversion unit 112 performs MTF correction on the target pixel by convolving and integrating the real space correction filter and the image data of the target pixel (S207).

  Next, the image output unit 113 determines whether or not the MTF correction has been completed for all the pixels of the original image (S208), and if there is an unprocessed pixel, the process returns to step S202, and the processes of steps S202 to S207 are performed. repeat. When the MTF correction for all the pixels is completed, the image data after MTF correction stored in a memory (not shown) is output to the image output device 103 (S209).

Feature Quantity Calculation Unit Details of the process (S203) of the feature quantity calculation unit 104 will be described with reference to the flowchart of FIG.

  The average luminance acquisition unit 107 calculates an average luminance value as a feature amount from the values of a plurality of pixels in a predetermined region including the target pixel (S301). The frequency conversion unit 105 decomposes the image data of the target pixel into frequency components by wavelet conversion or the like (S302). The amplitude acquisition unit 106 calculates an amplitude value as a feature amount for each frequency component from the pixel value corresponding to the target pixel using the image data decomposed into frequency components (S303).

● MTF acquisition department
The MTF storage unit 108 stores MTF data corresponding to a plurality of combinations of amplitude values and luminance values for each frequency component. Therefore, the MTF acquisition unit 109 acquires MTF data corresponding to the amplitude value and the average luminance value calculated for each frequency component by the feature amount calculation unit 104, and acquires MTF data of all frequency components.

Correction Filter Creation Unit Image quality degradation due to MTF of the image output apparatus 103 is expressed by the following equation as multiplication of image data to be output and MTF in the frequency space.
G (u, v) = H (u, v) × F (u, v)… (1)
Where G (u, v) is the output image (degraded image),
H (u, v) is MTF (deterioration function),
F (u, v) is the original image,
(u, v) is the spatial frequency.

As a method for recovering image quality deterioration, an image recovery method using a recovery filter is known. Examples of the recovery filter include an inverse filter of a degradation function and a Wiener filter. The following equation shows an inverse filter that is an inverse function I (u, v) of the degradation function H (u, v).
I (u, v) = 1 / H (u, v)… (2)

Since the inverse filter cannot take noise into consideration, noise components are emphasized when noise exists. On the other hand, the Wiener filter is a recovery filter that controls noise.
W (u, v) = Hc (u, v) / (| H (u, v) | ² 2 + Γ)… (3)
Where W (u, v) is the Wiener filter,
Hc (u, v) is the complex conjugate of H (u, v),
Г is a parameter that controls the noise component.

  The correction filter creation unit 110 creates an inverse filter or a Wiener filter as a recovery filter using the MTF corresponding to the feature amount as the deterioration function. However, the form of the recovery filter is not limited as long as it can recover or suppress deterioration in image quality.

Further, there are the following methods for creating a two-dimensional MTF correction filter using a one-dimensional MTF correction filter coefficient, but the present invention is not limited thereto.
・ A method to rotate the one-dimensional filter coefficient around the origin,
-For example, a method of interpolating between two directions using one-dimensional filter coefficients in two orthogonal directions,
A method of interpolating between a plurality of directions using one-dimensional filter coefficients of three or more directions.

● MTF correction When the degree of image quality degradation due to MTF of the image output device 103 depends on the feature value such as the amplitude value and average luminance value for each frequency component of the pixel, the degree of image quality degradation differs for each pixel. It is necessary to change the degree for each pixel. In order to change the degree of correction for each pixel, filtering in the real space is required, and the correction filter conversion unit 111 converts the correction filter in the frequency space into a correction filter in the real space.

The filtering in the real space is expressed as a convolution integral between the image of interest and the correction filter. Specifically, the restoration process by MTF correction in the real space is modeled as the following equation. The image conversion unit 112 performs MTF correction on the image input by the image input unit 101 according to the following equation.
g (x, y) = h (x, y) * f (x, y)… (4)
Where g (x, y) is the image after MTF correction,
h (x, y) is a correction filter for MTF correction,
f (x, y) is the image before MTF correction,
* Represents convolution integral.

  The image processing according to the second embodiment of the present invention will be described below. Note that the same reference numerals in the second embodiment denote the same parts as in the first embodiment, and a detailed description thereof will be omitted.

  A configuration example of the image processing apparatus according to the second embodiment will be described with reference to the block diagram of FIG.

  The image processing apparatus 100 according to the second embodiment is connected to a measurement unit 122 that measures the density distribution of the MTF measurement chart output from the image output apparatus 103. As the measurement unit 122, a scanner, a digital camera, or the like can be used. The image processing apparatus 100 according to the second embodiment includes an MTF calculation unit 121. The MTF calculation unit 121 calculates MTF data of the image output apparatus 103 from the density distribution input from the measurement unit 122, and stores the calculated MTF data in the MTF storage unit.

  The calculation of MTF data will be described with reference to the flowchart of FIG.

  The image output unit 113 outputs the MTF measurement chart image data stored in a memory (not shown) to the image output apparatus 103, and causes the image output apparatus 103 to output the chart (S501). The MTF calculation unit 121 controls the measurement unit 122 to measure the chart, and inputs the concentration distribution data of the measurement result (S502). Then, MTF data of the image output device 103 is calculated from the density distribution data (S503), and the calculated MTF data is stored in the MTF storage unit 108 (S504).

Measurement of MTF There are the following methods for measuring the MTF of the image output apparatus 103. (1) A method in which a sine wave chart is output to the image output device 103 as a chart for MTF measurement, and the MTF is calculated from the maximum density value and the minimum density value read from the sine wave chart. (2) A rectangular wave chart is output to the image output device 103 as a chart for MTF measurement. Then, a contrast transfer function (CTF) is calculated from the maximum density value and the minimum density value read from the rectangular wave chart, and CFT is converted to MTF using Coltman's correction formula. The MTF measurement method is not particularly limited.

  In the second embodiment, the MTF of the image output device 103 is measured using a chart created using a plurality of frequency parameters, amplitude parameters, and average luminance parameters as a sine wave chart or a rectangular wave chart.

  An example of the MTF measurement chart will be described with reference to FIGS. The chart shown in FIG. 9 is a sine wave chart, and the average luminance value is equal and the amplitude value is different for each frequency. Therefore, if the sine wave chart shown in FIG. 9 is used, the change in MTF with respect to the change in amplitude value can be measured.

  The chart shown in FIG. 10 is also a sine wave chart, in which the amplitude value is equal and the average luminance value is different for each frequency. Therefore, if the sine wave chart shown in FIG. 10 is used, the change in MTF with respect to the change in average luminance value can be measured.

  The MTF calculation unit 121 inputs a chart image obtained by converting the MTF measurement chart output from the image output device 103 into data using an image input device (measurement unit 122) such as a scanner or a digital camera. Then, MTF data is calculated from the density distribution (maximum density value and minimum density value) of the chart image. At that time, if the MTF of the image input device is canceled from the calculated MTF data, the MTF data of the image output device 103 alone can be obtained.

  Hereinafter, image processing according to the third embodiment of the present invention will be described. Note that the same reference numerals in the third embodiment denote the same parts as in the first and second embodiments, and a detailed description thereof will be omitted.

[Device configuration]
A configuration example of the image processing apparatus according to the third embodiment will be described with reference to the block diagram of FIG.

  The image processing apparatus 1100 is connected to an image output apparatus 1103 such as an ink jet printer or an electrophotographic printer, and an image storage unit 1102 for storing images. Note that the image storage unit 1102 may be a server device connected via a network. The image input unit 1101 inputs image data of an image (hereinafter referred to as an original image) to be output by the image output device 1103 from the image storage unit 1102.

  The feature amount calculation unit 1104 has the following configuration to calculate the feature amount of the target pixel of the image to be output. The frequency conversion unit 1105 decomposes the image data into frequency components using wavelet transform or the like. The amplitude acquisition unit 1106 acquires an amplitude value as a feature amount for each frequency component from the image data decomposed into frequency components using a pixel value corresponding to the target pixel. The average luminance acquisition unit 1107 calculates an average luminance value as a feature amount from the pixel value of a predetermined image area including the eye pixel.

  The luminance change amount storage unit 1108 stores the luminance change amount of the image output device 1103 corresponding to the image feature amount for each frequency component. The correction coefficient selection unit 1109 inputs the feature amount (amplitude value and average luminance value) of the target pixel. Then, the luminance change amount to be applied to the target pixel is selected as the luminance correction coefficient from the luminance change amount for each frequency component stored in the luminance change amount storage unit 1108.

  The luminance correction unit 1112 corrects the luminance of the image data of the image to be output using the luminance correction coefficient. The image output unit 1113 outputs the brightness-corrected image data to the image output device 1103, and prints the image.

[Image processing]
The brightness correction will be described with reference to the flowchart of FIG.

  The image input unit 1101 inputs image data of the original image (S1201). The feature amount calculation unit 1104 selects a luminance correction target pixel (target pixel) from the input image data (S1202). The frequency conversion unit 1105 decomposes the image data of the pixel of interest into frequency components by a method of decomposing into frequency components such as wavelet transform (S1203). The amplitude acquisition unit 1106 calculates an amplitude value as a feature amount for each frequency component from the pixel value corresponding to the target pixel using the image data decomposed into frequency components (S1204). Further, the average luminance acquisition unit 1107 calculates an average luminance value as a feature amount from the values of a plurality of pixels in a predetermined area including the target pixel (S1205).

  Next, the correction coefficient selection unit 1109 displays the luminance change amount for each frequency component of the image output device 103 according to the calculated feature amount (the amplitude value for each frequency component and the average luminance value) as the luminance change amount storage unit 1108. (S1206). Then, the luminance change amount used as the luminance correction coefficient corresponding to the feature amount of the target pixel is selected from the acquired luminance change amount for each frequency component (S1207).

  Next, as will be described in detail later, the luminance correction unit 1112 corrects the luminance of the image data of the target pixel using the luminance correction coefficient (S1208). The image output unit 1113 determines whether or not luminance correction has been completed for all the pixels of the original image (S1209). If there is an unprocessed pixel, the process returns to step S1202, and the processes of steps S1202 to S1208 are repeated. When the luminance correction for all the pixels is completed, the image data after luminance correction stored in a memory (not shown) is output to the image output device 1103 (S1210).

Acquisition of Luminance Change A luminance change storage unit 1108 stores a luminance change corresponding to a plurality of combinations of amplitude values and luminance values for each frequency component. The amount of change in luminance is a frequency, amplitude value, and a sine wave chart in which the luminance value is set to predetermined conditions. It is calculated from the difference of the average luminance value of. An example of the luminance change amount stored in the luminance change amount storage unit 1108 will be described with reference to FIG. Since the feature amount of the pixel of interest is calculated for each frequency component, the luminance change amount corresponding to the amplitude value and the average luminance value is obtained by acquiring the luminance change amount corresponding to the amplitude value and the average luminance value for each frequency component. Can be obtained for all frequency components.

Selection of luminance change amount (luminance correction coefficient) The correction coefficient selection unit 1109 refers to the luminance change amount corresponding to the feature quantity for each frequency component, and the luminance value of the target pixel after the luminance change and before the luminance change The luminance change amount with the smallest difference in average luminance values is selected as the luminance correction coefficient. This difference is calculated by subtracting the product of the luminance change amount and the average luminance value from the average luminance value under the conditions of a certain frequency, amplitude value, and average luminance value. However, the method for selecting or calculating the luminance correction coefficient is that if the luminance correction coefficient that makes the luminance value of the output image equal to the luminance value of the original image can be calculated when correcting the luminance change in the output image, The method is not limited.

  Note that there may be a case where the luminance change amount corresponding to the feature amount of the target pixel does not exist in the discrete luminance change amount stored in the luminance change amount storage unit 1108. In that case, the correction coefficient selection unit 1109 performs an interpolation operation on the luminance change amount corresponding to the feature amount of the target pixel using the luminance change amount of the feature amount stored in the luminance change amount storage unit close to the feature amount of the target pixel. .

Luminance Correction Details of the luminance correction (S1208) of the target pixel will be described with reference to the flowchart of FIG.

  The luminance correction unit 1112 receives the image data of the target pixel (S1301), and inputs the luminance correction coefficient selected for each frequency component (S1302). Then, for each frequency component, the product of the luminance correction coefficient and the average luminance value of the target pixel is calculated, and the sum of these products is added as a correction value to the input image data of the target pixel (S1303). If the image data with the correction value added deviates from the range of the image data (for example, a range of 0 to 255 for an 8-bit image), the image data for the deviation is diffused to the unprocessed pixels around the target pixel. (S1304). That is, a clipping process for preserving luminance within a predetermined area is performed.

  The luminance correction method is not limited as long as the luminance of the image can be corrected. Similarly, the clipping process is not limited as long as the image data can be within the range of the image data.

  As described above, the luminance correction based on the luminance change characteristic of the image output apparatus can be adaptively performed for each pixel by using the luminance correction coefficient corresponding to the feature amount for each frequency component.

[Modification]
In the first and second embodiments, the MTF correction corresponding to the change in the MTF of the image output apparatus has been described. In the third embodiment, the luminance correction corresponding to the luminance change of the output image has been described. These MTF correction and luminance correction can be executed simultaneously. In that case, after creating a correction filter for MTF correction and selecting a luminance correction coefficient, the DC component of the correction filter is adjusted based on the luminance correction coefficient. Then, the adjusted correction filter is converted into a real space, and MTF correction and luminance correction are simultaneously performed on the image data of the target pixel by convolution integration. Other processes are substantially the same as those in the first and third embodiments, and detailed description thereof is omitted.

[Other Examples]
The present invention can also be realized by executing the following processing. That is, software (program) that realizes the functions of the above-described embodiments is supplied to a system or apparatus via a network or various storage media, and a computer (or CPU, MPU, etc.) of the system or apparatus reads the program. It is a process to be executed.

Claims (9)

Calculating means for calculating the feature amount of the image for each frequency component from the image data obtained by decomposing the input image data into frequency components, using a pixel value corresponding to the target pixel ;
Obtaining means for obtaining an optical transfer function corresponding to the feature amount for each frequency component from storage means for storing the optical transfer function of the image output device for each frequency component;
Creating means for creating a correction filter in a frequency space for correcting the input image from an optical transfer function acquired for each frequency component;
Converting means for converting the correction filter in the frequency space into a correction filter in the real space;
Correction means for performing correction for each pixel of the input image using the correction filter in the real space;
An image processing apparatus comprising: output means for outputting the corrected image to the image output apparatus.   2. The image processing apparatus according to claim 1, wherein the processing of the calculation unit, acquisition unit, creation unit, conversion unit, and correction unit is repeated for each pixel.   3. The image processing apparatus according to claim 1, wherein the correction unit performs the correction by convolution integration of a correction filter in the real space and a pixel value. From the image data in which the input image data is decomposed into frequency components, using the pixel value corresponding to the target pixel , calculate the feature amount of the image for each frequency component,
From the storage means for storing the optical transfer function of the image output device for each frequency component, for each frequency component, obtain an optical transfer function according to the feature amount,
Create a frequency space correction filter that corrects the input image from the optical transfer function acquired for each frequency component,
Converting the frequency space correction filter into a real space correction filter;
Using the real space correction filter, correct each pixel of the input image,
An image processing method, comprising: outputting the corrected image to the image output device. Calculating means for calculating the feature amount of the image for each frequency component from the image data obtained by decomposing the input image data into frequency components, using a pixel value corresponding to the target pixel ;
An acquisition unit that acquires a luminance change amount corresponding to the feature amount for each frequency component from a storage unit that stores the luminance change amount of the output image by the image output device for each frequency component;
A selection means for selecting a luminance change amount as a luminance correction coefficient from the acquired luminance change amount;
Correction means for correcting the luminance for each pixel of the input image using the luminance correction coefficient;
An image processing apparatus comprising: output means for outputting the corrected image to the image output apparatus. 6. The image processing apparatus according to claim 5 , wherein the processing of the calculation unit, the acquisition unit, the selection unit, and the correction unit is repeated for each pixel. From the image data in which the input image data is decomposed into frequency components, using the pixel value corresponding to the target pixel , calculate the feature amount of the image for each frequency component,
From the storage means for storing the luminance change amount of the output image by the image output device for each frequency component, obtain the luminance change amount according to the feature amount for each frequency component,
Select a luminance change amount as a luminance correction coefficient from the acquired luminance change amount,
Using the brightness correction coefficient, correct the brightness for each pixel of the input image,
An image processing method, comprising: outputting the corrected image to the image output device. A program for causing a computer to function as each unit of the image processing apparatus according to any one of claims 1 to 3, 5, and 6 . 9. A computer-readable recording medium on which the program according to claim 8 is recorded.