JP2003283878A - Image quality improvement method - Google Patents

Abstract

(57) Abstract: An object of the present invention is to provide a new image quality improving method for improving image quality by performing good noise removal on an image on which image quality correction processing is performed. A process of correcting a level value of an image, a process of determining a noise removal strength based on a level value correction amount in a portion of the image to which the noise removal process is applied or a portion in the vicinity thereof, and a process of determining the determined noise According to the removal strength
And a step of removing noise for each part of the image. Then, in addition to this configuration, a configuration including a process of evaluating the noise intensity included in the image before correction may be adopted, and processing may be performed to determine the noise removal intensity with reference to the evaluated noise intensity. .

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image quality improving method for improving image quality by removing image noise, and more particularly, to adaptively removing image noise when correcting a level value of an image. To improve the image quality.

[0002]

2. Description of the Related Art The following techniques are known as conventional techniques for adaptively determining the image noise removal intensity to remove image noise.

(1) Image processing apparatus disclosed in Japanese Patent Laid-Open No. 11-32236: In the image processing apparatus disclosed in Japanese Patent Laid-Open No. 11-32236,
Generate a histogram from the gradation distribution, and from its shape,
While controlling the amount of brightness correction by correction,
The strength of noise removal is adjusted.

FIG. 7 shows the configuration of the image processing apparatus disclosed in Japanese Patent Laid-Open No. 11-32236. In the figure, the low-pass filter is the noise removing means.

(2) Contour correction circuit disclosed in JP-A-5-64038 In the contour correction circuit disclosed in JP-A-5-64038,
The noise removal strength is adjusted according to the average brightness of the image.

FIG. 8 shows a device configuration of the contour correction circuit disclosed in Japanese Patent Laid-Open No. 5-64038. In the figure, YN
R is a noise removing means, and Beam detection is a means for obtaining an average brightness.

(3) Noise elimination device for video camera disclosed in Japanese Unexamined Patent Publication No. 8-18852, a noise elimination device for video camera disclosed in Japanese Unexamined Patent Publication No. 8-18852 discloses an exposure amount control signal (AGC) for the camera. The gain) is used to adjust the strength of noise removal.

FIG. 9 shows a device configuration of a noise removing device for a video camera disclosed in Japanese Patent Laid-Open No. 18852/1996.

In the figure, a YNR controller and a CNR controller are means for determining the AGC gain.

(4) Noise reduction device disclosed in Japanese Unexamined Patent Publication No. 8-149343: In the noise reduction device disclosed in Japanese Unexamined Patent Publication No. 8-149343,
The intensity of noise removal is adjusted for each pixel according to the level value of the pixel.

FIG. 10 shows the configuration of the noise reduction device disclosed in Japanese Patent Laid-Open No. 8-149343.

[0012]

According to the above conventional technique, it is possible to remove noise depending on the state of the image. However, there are the following problems.

In the image processing technique, in general, various correction processes are performed on the image in order to improve the image quality (brightness, contrast, etc.) of the image.

For example, in the correction for increasing the contrast,
A correction process using an S-shaped correction curve as shown in FIG. 11 is performed. Further, in the correction for suppressing the contrast, a correction process using an inverse S-shaped correction curve as shown in FIG. 12 is performed.

When a correction process using such a correction curve is performed, the noise existing in the image before the correction process is
Areas where the inclination of the correction curve is steep are enlarged, and areas where the inclination of the correction curve is gentle are reduced.

However, the above-mentioned conventional techniques (1)-
In (3), the noise removal intensity is adjusted for different images, but the same intensity noise removal is performed for one image regardless of the image part. ing.

Therefore, if the noise removal strength is set on the basis of the area where the noise is enlarged, there is a problem that the noise is removed with excessive strength for the area where the noise is reduced.

As pointed out in the above-mentioned conventional technique (3) and the like, excessive noise removal impairs the sharpness of an image, which is not preferable.

On the contrary, when the noise removal strength is set on the basis of the area where the noise is reduced, the strength becomes insufficient for the area where the noise is enlarged, and as a result, the noise appears. is there.

In the above-mentioned prior art (4), the noise removal strength is determined based only on the image level value, and the noise removal strength is determined regardless of the image quality correction amount.

As a result, when the correction for increasing the contrast as shown in FIG. 11 is performed, the range of the image before correction is greatly expanded after the correction in the area of medium brightness shown in FIG. The noise having the amplitude of 10 will be expanded to the amplitude of 20 after the correction.

On the contrary, when the correction for lowering the contrast as shown in FIG. 12 is performed, in the illustrated area of medium brightness,
Since the range of the image before the correction is reduced after the correction, for example, the noise having the amplitude of 10 before the correction is reduced to the amplitude of 5 after the correction.

Even in such a case, in the prior art (4), since the noise removal strength is determined only on the basis of the image level value, the noise removal becomes insufficient when the contrast is increased. However, there is a problem that the noise removal becomes excessive when the correction for reducing the contrast is performed.

As described above, in the prior art, the noise removal by a constant value or the noise removal according to the image level value is performed on the entire area.

Therefore, according to the prior art, it is possible to avoid the problem that an image is subjected to image quality correction processing of different sizes depending on the part, which causes insufficient noise removal or excessive noise removal depending on the part. There is a problem that you cannot do it.

The present invention has been made in view of the above circumstances, and a new image quality is improved by performing good noise removal on an image for which image quality correction processing such as contrast correction and brightness correction is performed. The purpose is to provide an improvement method.

[0027]

In order to achieve this object, the image quality improving method of the present invention corrects the level value of the image when the image quality is improved by removing the noise of the image. The process and the process of determining the noise removal strength based on the level value correction amount in the part of the image to which the noise removal processing is applied or the vicinity thereof, and the noise in the part of the image according to the determined noise removal strength And a step of removing.

At the time of adopting this configuration,
The method may include a step of evaluating the noise intensity included in the image before correction. At this time, in the step of determining the noise removal intensity, processing is performed so as to determine the noise removal intensity by also referring to the evaluated noise intensity. It will be.

Each of the above processing steps can be realized by a computer program, and this computer program can be provided by being recorded in a recording medium such as a semiconductor memory.

Further, the image quality improving method of the present invention having this configuration, based on the means for correcting the level value of the image and the level value correction amount at the part of the image to which the noise removal processing is applied or the part in the vicinity thereof, The image quality of the present invention, which comprises a unit for determining the noise removal strength, a unit for removing noise in units of image parts in accordance with the determined noise removal intensity, and a unit for evaluating the noise intensity included in the image before correction. It will be realized by the improvement device.

According to this configuration, according to the present invention, it becomes possible to adaptively determine the noise removal strength for an image on which image quality correction processing such as contrast correction and brightness correction is performed, and thus, the image is corrected. However, even when various level value correction processing is performed, it is possible to generate an image with less noise by performing good noise removal.

Next, the basic processing flow of the present invention will be described with reference to FIGS.

FIG. 1 shows a process for determining the noise removal strength with reference to the level value correction amount, and FIG. 2 shows the noise removal strength with reference to the level value correction amount and the noise strength. The processing when it does is shown. As shown in this figure, the present invention proceeds with each image.

In the present invention, as described in the section "Problems to be solved by the invention", image quality correction processing for correcting the image level value such as contrast correction and brightness correction (hereinafter, simply referred to as "image quality correction processing"). The noise removal process is performed when performing the (.

In the present invention, as shown in FIG. 1, when determining the intensity of noise removal, the correction amount of this level value is referred to, and the intensity of noise removal is determined based on this amount.
Therefore, the noise removal strength can be increased for the portion where the noise is enlarged (potentially) by the image quality correction, and the effective noise removal can be performed.

Here, in FIG. 1, the blocks of the image quality correction processing are not explicitly shown. The reason is that the image quality correction process may be performed at various stages. For example, before the processing of FIG. 1 is started, correction may be collectively performed in advance, and then the processing of FIG. 1 may be performed.
"Refer to the correction amount for the selected part or its vicinity"
You may go at the same time (immediately before or after).

Although the processing of FIGS. 1 and 2 is executed for each part of the image, it is effective to proceed the processing for each pixel, for example, because the entire image can be processed by a simple scan. .

Further, in determining the noise removal strength, it is also effective to refer to not only the correction amount but also the noise appearance level (magnitude) of the selected portion before correction, as shown in FIG. Target. For example, when noise is likely to occur near the edge, the presence or absence of the edge may be evaluated and the noise intensity may be estimated based on the result. These estimation results are also effective when referred to when determining the noise removal strength.

In addition, if the intensity of noise changes depending on the level value (usually equivalent to brightness or brightness) (for example, noise is likely to occur in the dark), a table or a mathematical formula is used in advance. It is also possible to define the correspondence between the level value and the noise intensity and estimate the noise intensity by referring to this table or mathematical formulas.

It is also effective to use the absolute value of the correction amount for the level value correction amount, which is used as a reference when determining the noise removal strength. For example, there is a clipping process as shown in FIG. 3 as a correction for making the highlight part clear.

In this case, since the dark area has a small correction amount, the change in color before and after the correction is small, and the change in the conspicuousness of noise is not so different before and after the correction. However, since the bright part has a large correction amount, there is a possibility that the conspicuousness of noise may significantly change. In other words, by using the absolute value of the correction amount as a reference, it is possible to increase the noise removal strength at a place where the noticeability of noise is likely to change, and effective noise removal can be performed.

In contrast correction as shown in FIG. 11 and FIG. 12, the observability of noise after correction is increased (or decreased) depending on the change amount of the level value correction amount with respect to the minute change of the level value by the image quality correction processing. ) Can be estimated, and it is also effective to use such a value as a reference. The ratio of the change width of the input data and the change width of the output data is desirable as the evaluation criterion for such a change amount. The ratio may be obtained by a method substantially equivalent to a differential coefficient which is a value in a minute section.

The noise intensity estimation, the absolute value of the correction level, and the change amount of the level value correction amount with respect to a minute change in the level value are preferably evaluated in accordance with the human sense, and therefore, they are performed in a uniform color space such as CIELAB. Is effective.

[0044]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, the present invention will be described in detail according to embodiments.

FIG. 4 shows an apparatus configuration of the digital still camera 1 having the present invention.

In this figure, among the functions of the digital still camera 1, only the portions related to the image processing related to the present invention are shown, and the functions related to motor drive which can be ignored in the description of the present invention. Functions not related to image processing, such as, are not shown.

Digital still camera 1 having the present invention
Has a function of automatically correcting the brightness of an image (γ correction function) and a function of removing image noise.

As shown in this figure, in the digital still camera 1 equipped with the present invention, the image signal picked up by the CCD 10 is A / D converted by the A / D converter 11 and input to the arithmetic unit 12. Γ correction will be performed.

The γ correction is a correction by the formula S _output = S _input ^r , which changes the gradation characteristics of the image and is used, for example, when changing the brightness.

Here, S _input represents the signal level before correction (assumed to be normalized in the range of 1 to 0), and S _output represents the signal level after correction.

In the following description, "γ" shown in the equation may be represented by "r" for convenience of description.

The arithmetic unit 12 has a CPU for executing processing according to a code stored in the ROM 13 and a JPEG.
Image Processing that performs standard image processing such as compression
It is composed of Unit and.

The arithmetic unit 12 appropriately writes data in the RAM 14 and reads the data from the RAM 14 to proceed with the processing. The image data JPEG-compressed by the arithmetic unit 12 is stored in the non-volatile storage medium 15.

FIG. 5 shows a flowchart of the image quality improving process according to the present invention, which is executed by the CPU of the arithmetic unit 12.

The image picked up by the CCD 10 is once R
It shall be stored in AM14. In this embodiment example,
Before this storage, general γ correction (eg γ = 1 /
2.2) shall be performed and the data range shall be set so that the values should be distributed between 0 and 255.

The stored image is input as the image to be processed, and the process is executed according to the flowchart shown in FIG.

First, in step 1, an image to be processed is input. Then, in step 2, the γ value required to perform the correction (γ correction) such that the brightness of the image becomes appropriate (the general γ correction described above is followed by further correction Γ value) for adjusting the appropriateness.

The calculation process of the γ value is performed by calculating a γ value larger than 1 if the image is too bright and calculating a γ value smaller than 1 if the image is too dark. . For example, the document “Juha Katajamaki and Pekka Laihan
en, "Image Dependent Gamma Selection Based on Colo
r Palette Equalization and a Simple Lightness Mode
l ", Proc. of 7th CIC, 301-306 (1999)".

Subsequently, in step 3, a process for removing noise from the entire image is started. In this example of the embodiment, the process proceeds in units of pixels.

That is, starting from the upper left pixel as the starting point (step 3), the pixels are sequentially selected from the upper left to the lower right (step 9). First, at step 4, the selected pixel is selected at step 2 The correction amount of the level value when the correction is performed by the obtained γ value is calculated.

For example, R _correct = [(R _input / 255) ^r ] × 255-R
_input G _correct = [(G _input / 255) ^r ] × 255-G
_input B _correct = [(B _input / 255) ^r ] × 255-B
The correction amount of the level value is calculated according to the formula called _input .

Here, R _correct is a correction amount, and R _{correct
input} is the R value of the correction target before correction. The same notation is used for G and B.

Then, in step 5, the noise removal strength is calculated using this correction amount. Here, the larger the amount of correction by γ correction, the greater the change in color due to correction, so the change in the appearance of noise is also large, and it is determined that there is a high risk of noise being magnified and the noise removal strength is determined. Make stronger

For example, [Formula 1] R _remove-power = | R _correct | /20+0.5 if R _remove-power > 1 then R _remove-power = 1 G _remove-power = | G _correct | /20+0.5 if G _remove-power > 1 then G _remove-power = 1 B _remove-power = | B _correct | /20+0.5 if B _remove-power > 1 then B _remove-power = 1 (hereinafter referred to as formula 1) Then, the strength of denoising is calculated.

Here, R _remove-power is the noise removal strength for the R value, and G and B follow the same notation method.

In this equation 1, noise removal with a removal intensity of 0.5 or more is performed on all pixels, and regarding the pixels determined to be required to remove noise strongly according to the above-mentioned criteria, , The removal strength is set to a value close to 1 (1 at the maximum).

Subsequently, in step 6, noise removal is executed according to the determined noise removal strength. There are various methods such as LPF as a method of removing noise, but here, a median filter (pixel value of the target pixel is set to a pixel value of an intermediate size included in the pixel values of neighboring pixels, which is known in this technical field. Noise removal processing is performed by a replacement filter operation). For example, the filter size may be 3 × 3 pixels.

Regarding the median filter, "Ch
ristopher Watkins, Alberto Sadunand Stephen Marenk
a, "Modern Image Processing", Academic Press, Inc.
(1993) pp. 64-65 ".

For example, [Formula 2] R _{noize-removed} = R _remove-power × R _median + (1-R _remove-power ) × R _input G _{noize-removed} = G _remove-power × G _median + (1-G _remove-power ) × G _input B _{noize-removed} = B _remove-power × B _median + (1-B _remove-power ) × B _{input according to} the formula (hereinafter, may be referred to as Formula 2) To do.

Here, R _{noize-removed} is the noise removal result, and R _median is the median filter application result (replaced pixel value). The same notation is used for G and B.

As can be seen from this equation, R_remove-power
Is larger, the noise removal result is R _medianValue close to
R, G_remove-powerIs larger, the noise removal result is G
_{medi an}Becomes a value close to_remove-powerIs larger,
The noise removal result is B_medianIt is a value close to.

Subsequently, in step 7, the image quality is corrected by executing the γ correction, and in the following step 8, it is judged whether or not the processing is completed for all the pixels, and the processing is completed for all the pixels. If it is determined that the processing is not completed, the process proceeds to step 9, and the next pixel is selected. On the other hand, if it is determined that the processing is completed for all the pixels, the processing is ended.

As described above, in the digital still camera 1 including the present invention, it is possible to focus noise removal on a place where the correction amount by the γ correction is large, and prevent the noise expansion by the γ correction. become able to.

Furthermore, it is also effective to make the following modifications to this embodiment.

(1) Modified Example 1 In the above-described embodiment, the amount of correction of the image level value is calculated using RGB values, but in a uniform color space such as CIELAB (LAB color space defined by CIE). It is also effective to do.

In this case, Expression 1 (calculation expression for noise removal strength) described above is replaced with Expression 3 described below, and Expression 2 (calculation expression for executing noise removal) described above.
May be replaced with the following Equation 4.

RGB values before γ correction (R _input ,
B _input , G _input ) and the RGB value (R
_corrected = R _input + R _correct , G _corrected = G
_input + G _{correc t} , B _corrected = B _input + B
_correct ) and, for example, according to the definition of sRGB, CIELAB values (L ^* _input , a ^*)
_input , b ^* _input ) and CIELAB value (L ^*
_corrected , a ^* _corrected , b ^* _corrected ). This conversion is a well-known method widely used in the field of image processing.

[0078] Then, according to Equation 3 shown below, for example, [Equation 3] CIELAB _remove-power = ^{_{[((L * corrected -L * input}} ) 2 + (a * corrected -a * input) 2 + (b * _corrected -b ^* _input ) ² ) ^1/2 ] / 10 if CIELAB _remove-power > 1 then CIELAB _remove-power = 1 such that “1/10” of the difference value of CIELAB before and after γ correction is noise. It is calculated as the removal strength.

On the other hand, for noise removal, for example, according to Equation 4 below, [Equation 4] R _{noize-removed} = CIELAB _remove-power × R _median + (1-CIELAB _remove-power ) × R _input G _{noize- removed} = CIELAB _remove-power × G _median + (1-CIELAB _remove-power ) × G _input B _{noize-removed} = CIELAB _remove-power × B _median + (1-CIELAB _remove-power ) × B _input .

(2) Modification 2 In the above-described embodiment, the noise removal strength calculation formula shown in Formula 1 is calculated using the absolute value of the correction amount, but not the absolute value, but the level value. The calculation may be performed by the change amount (corresponding to the differential coefficient) of the level value correction amount with respect to the minute change of.

Since the relationship of the amount of change with respect to this minute change corresponds to the relationship of the amplitude of the noise of the image before and after the image quality correction processing, it can be said that it is one of the indexes showing the change in the visibility of the noise due to the image quality correction. If this value is large, the noise removal strength is increased.

For example, R _{remove-power-correct} = (γ / 2) x (R _input / 255) ^r-1 if R _{remove-power-correct} > 1 then R _{remove-power1correct} = 1 G _{remove-power-correct} = (Γ / 2) × (G _input / 255) ^r-1 if G _{remove-power-correct} > 1 then G _{remove-power-correct} = 1 B _{remove-power-correct} = (γ / 2) × (B _input / 255) ^r-1 if B _{remove-power-correct} > 1 then B _{remove-power-correct} = 1, so that the noise removal strength is calculated by multiplying the differential coefficient of the γ-correction formula by “1/2”. To do.

(3) Modification 3 In the above-described embodiment, the noise removal strength is calculated based on the correction amount of the image quality correction. In addition to this, the noise strength (estimation) of the removed portion is further calculated. Value) may be added. Specifically, when noise stands out near the edge, the noise intensity is calculated based on the edge intensity.

In order to realize this, the flow chart shown in FIG. 5 is changed to that shown in FIG.

That is, a step of calculating the noise intensity (step α) is added between step 4 and step 5,
By changing the calculation formula of the noise removal strength as shown below, the noise removal strength of the portion where the noise is estimated to be strong can be increased, and the noise can be satisfactorily removed.

Here, it is assumed that the edge detection is performed by a publicly known technique of a 3 × 3 pixel Sobel filter (a filter operation for obtaining a Sobel filter applied value (edge strength) showing a large value for a pixel located at an edge). To do.

Regarding the Sobel filter, "Chri
stopher Watkins, Alberto Sadun and Stephen Marenk
a, "Modern Image Processing", Academic Press, Inc.
(1993) pp. 59-60 ".

Noise is basically the pixel of an edge.
It is not a thing, but it has the property that it stands out near the edge.
To do. Therefore, the result of applying the Sobel filter for the pixel to be processed
(R _sobel, G_sobel, B_sobel) And its vicinity (eg
Sobel filter application result in 5 x 5 pixel range)
Maximum value (R_sobel-max, G_sobel-max，
B_{sobel-ma x}) And the maximum pixel edge strength
If the value is larger, the target pixel is near the edge.
So that the noise removal strength is strengthened.
To do.

For example, The amount of enhancing the noise removal strength is calculated according to the following equation.

For concrete noise removal strength, for example,
For example, it is calculated in this way (R _edge, G_edge, B
_edge) Is used to describe in Equation 1.
(R_remove-power, G_{re move-power}, B_remove-power)
To [Formula 5] R_{remove-power-new}= (R_remove-power+ R_edge) / 2 G_{remove-power-new}= (G_remove-power+ G_edge) / 2 B_{remove-power-new}= (B_remove-power+ B_edge) / 2 According to the formula (hereinafter, sometimes referred to as Formula 5)
, (R_{remove-power-new}, G_{remove-power-new}, B
_{remove-power-new}) Noise removal strength
And use the noise removal strength thus obtained
Noise will be removed.

Alternatively, it is calculated in this way (R
_edge , G _edge , B _edge ) and describe it in Formula 1 (R _remove-power , G _remove-power , B _remove-power )
[Equation 6] R _{remove-power-new} = (R _remove-power + R _edge ) if R _{remove-power-new} > 1 then R _{remove-power-new} = 1 G _{remove-power-new} = (G _{remove- power} + G _edge ) if G _{remove-power-new} > 1 then G _{remove-power-new} = 1 B _{remove-power-new} = (B _remove-power + B _edge ) if B _{remove-power-new} > 1 then B _{remove- According to} the formula _power-new = 1 (hereinafter, may be referred to as Formula 6), (R _{remove-power-new} , G _{remove-power-new} , B
_{remove-power-new} ) to obtain the noise removal strength, and the noise removal strength thus obtained may be used to perform the noise removal.

Further, as a similar example to this modification 3, when noise is noticeable in a dark part, a coefficient for evaluating the darkness is obtained, and the noise removal strength is calculated according to the same method as in the equations (5) and (6). You may change it.

Specifically, for example, for each of the RGB values, the correspondence between the level value and the evaluation coefficient is tabulated and stored in the ROM 13, and the noise removal strength is changed with reference to this value. For example, when R = 0, the coefficient value “64/64”, when R = 1, the coefficient value “63/64”, R
= 2, the coefficient removal value is changed by referring to the coefficient value "62/64", ..., And the coefficient value "0/64" when R> 64.

Further, when noise is noticeable in an achromatic color, a similar table is created according to the saturation, and the noise removal strength is changed with reference to this value.

The embodiments of the present invention and the modifications thereof have been described above. Here, the median filter is used for the noise removal processing, but other methods may be used. For example, simple average or LPF may be used.

Further, although the noise removal strength is set for each pixel, it may be set by referring to the values of surrounding pixels (for example, calculating a value averaged by several pixels). Because the noise is
This is because it becomes apparent in a relative relationship with surrounding pixels.

Further, as the image quality correction processing, γ correction is taken as an example, but it may be used together with other correction such as dynamic range correction.

The correction amount may not be calculated but may be set by the user using some kind of user interface.

(Supplementary Note 1) An image quality improving method for improving image quality by removing noise from an image, in which a process of correcting a level value of an image and a portion of an image to which noise removal processing is applied or a vicinity thereof are performed. Image quality improvement comprising a step of determining a noise removal strength based on a level value correction amount in a part and a step of removing noise in units of an image part according to the determined noise removal strength. Method.

(Supplementary Note 2) The image quality improving method according to Supplementary Note 1 includes a step of evaluating the noise intensity included in the image before correction, and in the step of determining the noise removal intensity, the evaluated noise intensity is also referred to. A method of improving image quality, characterized in that the noise removal strength is determined by

(Supplementary Note 3) In the image quality improving method according to Supplementary Note 2, in the process of evaluating the noise intensity, the noise intensity is evaluated in a uniform color space.

(Supplementary Note 4) In the image quality improving method according to Supplementary Note 2 or 3, in the process of evaluating the noise intensity, the noise intensity is evaluated using the stored data defining the noise intensity associated with the level value of the image. Alternatively, an image quality improvement method characterized in that the noise intensity is evaluated using a prescribed mathematical formula.

(Supplementary Note 5) In the image quality improving method described in any one of Supplementary Notes 1 to 4, in the process of determining the noise removal strength, the noise removal strength is determined for each pixel. How to improve.

(Supplementary Note 6) In the image quality improving method according to any one of Supplementary Notes 1 to 5, in the process of determining the noise removal strength, the level at the part of the image to which the noise removal process is applied or a part in the vicinity thereof is applied. An image quality improving method characterized in that the noise removal strength is determined according to the absolute value of the value correction amount.

(Supplementary Note 7) In the image quality improving method according to any one of Supplementary Notes 1 to 5, in the process of determining the noise removal strength, a change in the level value correction amount with respect to a minute change in the level value due to the image quality correction processing is changed. An image quality improving method characterized in that the noise removal strength is determined according to the amount.

(Supplementary Note 8) In the image quality improving method according to Supplementary Note 6, in the process of determining the noise removal strength, the absolute value of the level value correction amount is evaluated in the uniform color space.

(Supplementary Note 9) In the image quality improving method according to Supplementary Note 7, in the process of determining the noise removal strength, the change amount of the level value correction amount is evaluated in the uniform color space.

(Supplementary Note 10) An image quality improving apparatus for improving image quality by removing noise from an image, comprising means for correcting the level value of the image, and a portion of the image to which noise removal processing is applied or a portion in the vicinity thereof. Image quality improvement comprising means for determining noise removal strength based on a level value correction amount in a region, and means for removing noise in units of image regions in accordance with the determined noise removal intensity apparatus.

(Supplementary Note 11) An image improvement program for performing a process for improving the image quality by removing noise from an image, which is a process for correcting a level value of an image and a portion of an image to which a noise removing process is applied. A method for causing a computer to execute a process of determining a noise removal intensity based on a level value correction amount in a region in the vicinity thereof and a process of removing noise in a unit of an image region according to the determined noise removal intensity Image improvement program.

(Supplementary Note 12) A recording medium of an image improvement program recording a program for performing a process for improving image quality by removing noise from an image, which is a process for correcting an image level value and a noise removing process. A process of determining the noise removal strength based on the level value correction amount in the region of the image to which is applied or a region in the vicinity thereof, and a process of removing noise in units of the image region according to the determined noise removal intensity. A recording medium for an image improvement program, which stores a program for causing a computer to execute.

[0111]

As described above, according to the present invention, the noise removal strength is determined based on the level value correction amount in the part of the image to which the noise removal process is applied or the part in the vicinity thereof, and the image is corrected accordingly. The noise is removed in units of.

According to the present invention, according to this configuration, even when various level value correction processing such as contrast correction is performed on an image, the image quality is improved by performing good noise removal. Will be possible.

[Brief description of drawings]

FIG. 1 is a diagram illustrating the principle of the present invention.

FIG. 2 is a diagram illustrating the principle of the present invention.

FIG. 3 is a diagram illustrating the principle of the present invention.

FIG. 4 is a device configuration diagram of a digital still camera including the present invention.

FIG. 5 is a flowchart of image quality improvement processing executed in the present invention.

FIG. 6 is a flowchart of image quality improvement processing executed in the present invention.

FIG. 7 is an explanatory diagram of a conventional technique.

FIG. 8 is an explanatory diagram of a conventional technique.

FIG. 9 is an explanatory diagram of a conventional technique.

FIG. 10 is an explanatory diagram of a conventional technique.

FIG. 11: S used in correction for increasing contrast
It is explanatory drawing of a character correction curve.

FIG. 12 is an explanatory diagram of an inverse S-shaped correction curve used for correction for suppressing contrast.

[Explanation of symbols]

1 Digital still camera 10 CCD 11 A / D converter 12 arithmetic unit 13 ROM 14 RAM 15 Non-volatile storage medium

Claims (5)

[Claims] 1. An image quality improving method for improving image quality by removing noise from an image, comprising: a step of correcting a level value of an image; and a portion of an image to which noise removal processing is applied or a portion in the vicinity thereof. An image quality improving method comprising: a step of determining a noise removal strength based on a level value correction amount; and a step of removing noise in units of image parts according to the determined noise removal strength. 2. The image quality improving method according to claim 1, further comprising a step of evaluating the noise intensity included in the image before correction, wherein the step of determining the noise removal intensity also refers to the evaluated noise intensity. An image quality improvement method characterized by determining the noise removal strength. 3. The image quality improving method according to claim 2, wherein, in the step of evaluating the noise intensity, the noise intensity is evaluated in a uniform color space. 4. The image quality improving method according to claim 1, wherein in the step of determining the noise removal strength, a level value at a site of an image to which the noise removal process is applied or a site in the vicinity thereof is applied. An image quality improvement method characterized in that the noise removal strength is determined according to the absolute value of the correction amount. 5. The image quality improving method according to claim 1, wherein in the process of determining the noise removal strength, a change amount of the level value correction amount with respect to a minute change of the level value by the image quality correction process. A method of improving image quality, characterized in that the noise removal strength is determined according to.