JP2011061419A - Image processor and image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology for simply determining whether a gradation value on which degrees are concentrated is the one in an area which should be noticed or the one of background, and performing ν conversion processing using an appropriate ν curve. <P>SOLUTION: The image processor has: a generation means for generating gradation histogram from an input image; a detection means for detecting the gradation value on which the degrees are concentrated from the gradation histogram to output the detected gradation value as a first gradation value; a calculation means for calculating the sum of the degrees within a near gradation range which is the gradation range for predetermined gradation including the first gradation value in the gradation histogram to output the calculated sum as a distribution amount; a correction means for outputting the first gradation value as it is when the distribution amount is equal to or less than a predetermined threshold, and correcting the first gradation value to be output when the distribution amount is larger than the predetermined threshold; a determination means for determining the ν curve according to the gradation value to be output by the correction means; and a ν conversion means for converting gradation of an image using the determined ν curve. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an image processing apparatus and an image processing method.

Conventionally, there is an image processing method that adaptively executes γ conversion processing according to the characteristics of an input image (input image; input image signal). Hereinafter, such an image processing method is referred to as dynamic γ processing.
As a method of dynamic γ processing, for example, there is a method of detecting a peak gradation value from a luminance histogram of an input image and adaptively executing γ conversion processing according to the frequency of the gradation value (Patent Document 1). reference). Further, there is a method in which a reference value is set using the frequency of the detected gradation value, and γ conversion processing is adaptively performed according to the reference value (see Patent Document 2).

Japanese Patent Laid-Open No. 06-350873 Japanese Patent Laid-Open No. 08-204963

However, in the conventional dynamic γ processing, as described below, an appropriate conversion result may not be obtained.
For example, assume that an image as shown in FIG. 2 is input. In FIG. 2, an area A is an object area (bright area) having a high gradation value as a main gradation value, and an area B is a flat background having a low gradation value as a main gradation value. It is an area (dark area). An example of a luminance histogram generated from such an image is shown in FIG.
As shown in FIG. 10, in the luminance histogram obtained from the image shown in FIG. 2, the frequencies are concentrated near the main gradation value a of the object and the main gradation value b of the background (tone value a And a peak appears in the gradation value b).
Here, in the image shown in FIG. 2, the area of the region B is larger than the area of the region A. Therefore, in the luminance histogram of FIG. 10, the peak of the gradation value b is more than the peak of the gradation value a. Get higher. Therefore, in the conventional method, the gradation value b is treated as a concentrated gradation (a gradation value of a region to be noted), and the gradation value is represented as a γ curve used for the gamma conversion process as indicated by a solid line in FIG. A γ curve that enhances the gradation of b and the vicinity thereof is set. A dotted line in FIG. 11 is a conversion straight line when the input signal is linearly converted into an output signal.
As a result, the background region B in the image of FIG. 2 can be displayed brightly as shown in FIG. 12 (the visibility of dark regions can be improved).

However, the region B is a flat background region and is not a notable region. Therefore, when γ conversion processing is performed using a γ curve as shown by the solid line in FIG. 11, the contrast between the object and the background is reduced, and the contrast of the image is impaired.
Originally, when an image as shown in FIG. 2 is input, as shown by the solid line in FIG. 6, it is preferable to set a γ curve that lowers the gradation value b and the gradation in the vicinity thereof. Thereby, as shown in FIG. 7, the contrast between the object and the background can be increased to express a good contrast. A dotted line in FIG. 6 is a conversion straight line when the input signal is linearly converted into an output signal.

Therefore, in the present invention, it is possible to easily determine whether the gradation value where the frequency is concentrated is the gradation value of the region to be noticed or the gradation value of the background, and γ conversion processing using an appropriate γ curve is performed. The purpose is to provide technology that can be performed.

  An image processing apparatus according to the present invention includes a generation unit that generates a gradation histogram from an input image, a detection unit that detects a gradation value in which frequencies are concentrated from the gradation histogram, and outputs the gradation value as a first gradation value. Calculating means for calculating the sum of the frequencies in the vicinity gradation range that is the gradation range for the predetermined gradation including the first gradation value in the gradation histogram, and outputting the distribution amount as a distribution amount; The first gradation value is output as it is when it is equal to or less than the threshold value, and the correction means for correcting and outputting the first gradation value when the distribution amount is larger than the predetermined threshold value, and the gradation output by the correction means A determining unit that determines a γ curve according to the value, and a γ converting unit that converts the gradation of the image using the determined γ curve, and is determined according to the corrected first gradation value The γ curve that is determined according to the first gradation value before correction is used. And a γ curve for obtaining an image having a low gradation in the vicinity gradation range as compared with an image whose gradation has been converted.

  The image processing method of the present invention includes a generation step of generating a gradation histogram from an input image, a detection step of detecting a gradation value at which frequencies are concentrated from the gradation histogram, and outputting the detected gradation value as a first gradation value. Calculating a total sum of frequencies in a neighboring gradation range that is a gradation range for a predetermined gradation including the first gradation value in the gradation histogram, and outputting the distribution amount as a distribution amount; When the value is equal to or smaller than the threshold value, the first gradation value is output as it is, and when the distribution amount is larger than the predetermined threshold value, a correction step for correcting and outputting the first gradation value, and a floor output in the correction step A determination step for determining a γ curve according to the tone value, and a γ conversion step for converting the gradation of the image using the determined γ curve, which is determined according to the corrected first gradation value Γ curve is the first gradation before correction Compared to the image gradation is transformed using γ curve is determined according to, characterized in that it is a γ curve for obtaining a gray scale having a low image near tonal range.

  According to the present invention, it is possible to easily determine whether the gradation value where the frequency is concentrated is the gradation value of the region of interest or the gradation value of the background, and γ conversion processing using an appropriate γ curve It can be performed.

FIG. 2 is a block diagram illustrating an example of a functional configuration of the image processing apparatus according to the first embodiment. The figure which shows an example of an input image. The figure which shows an example of a gradation histogram. The block diagram which shows an example of a function structure of a gradation value correction | amendment part. The figure which shows an example of the gamma curve selected by the gamma curve determination part. The figure which shows an example of (gamma) curve. The figure which shows an example of an output image. FIG. 9 is a block diagram illustrating an example of a functional configuration of an image processing apparatus according to a second embodiment. The figure which shows an example of a division area. The figure which shows an example of a gradation histogram. The figure which shows an example of (gamma) curve. The figure which shows an example of an output image.

<Example 1>
(overall structure)
Hereinafter, an image processing apparatus and an image processing method according to Embodiment 1 of the present invention will be described. FIG.
These are block diagrams which show an example of a function structure of the image processing apparatus which concerns on a present Example.
An image processing apparatus 100 illustrated in FIG. 1 includes a gradation histogram generation unit 101, a concentrated gradation detection unit 102, a distribution amount calculation unit 103, a gradation value correction unit 104, a γ curve determination unit 105, a γ conversion unit 106, and an input terminal. 107, an output terminal 108, and the like. Hereinafter, each function of the image processing apparatus 100 will be described in detail by taking as an example the case where the image shown in FIG. 2 is input.
In FIG. 2, an area A is an object area (bright area) having a high gradation value as a main gradation value, and an area B is a flat background having a low gradation value as a main gradation value. It is an area (dark area). In the present embodiment, the size of the image is 1920 pixels × 1080 pixels.

The gradation histogram generation unit 101 generates a gradation histogram from an input image (input image; input image signal).
In this embodiment, the gradation histogram generation unit 101 generates a gradation histogram (luminance histogram) of the luminance signal of the input image as the gradation histogram. Further, in this embodiment, as shown in FIG. 3, a luminance histogram of 256 categories with gradation values of 0 to 255 is generated. The horizontal axis indicates the gradation value, and the vertical axis indicates the frequency. The number of gradations and the number of categories are not limited to this. For example, the number of gradations may be 128, and the number of categories may be 64 (2 gradations per category).

The concentrated gradation detecting unit 102 detects a gradation value (concentrated gradation) where the frequency is concentrated from the gradation histogram generated by the gradation histogram generating unit 101, and outputs it as the first gradation value P1.
In this embodiment, the concentrated gradation detecting unit 102 detects the gradation value having the highest frequency as the concentrated gradation. More specifically, the gradation value having the highest frequency in the range of gradation values 0 to 63 is set as the first gradation value P1. The first gradation value P1 is not limited to the gradation value having the highest frequency. The first gradation value P1 may be a gradation value in which the frequency is concentrated. For example, the entire gradation range is divided into a plurality of gradation ranges (partial gradation ranges), the partial gradation range with the highest frequency is detected as the gradation range where the frequencies are concentrated, and the gradation value at the center is determined as the first gradation value. One gradation value P1 may be used. In the present embodiment, it is assumed that the gradation value 32 is set to the first gradation value P1 as shown in FIG.

  The distribution amount calculation unit 103 calculates the sum of the frequencies in the vicinity gradation range that is the gradation range for the predetermined gradation including the first gradation value P1 in the gradation histogram generated by the gradation histogram generation unit 101. Calculate and output as distribution amount S1. Specifically, the sum of the frequencies of the first gradation value P1 and the previous M gradation and the rear N gradation of the first gradation value P1 is defined as the distribution amount S1 (in this embodiment, the hatching in FIG. 3). (M = N = 2 as in the case). Note that the values of M and N are not limited to 2 (for example, M = N = 5, 10 may be used, or the values may be different between M and N). Further, the partial gradation range including the first gradation value P1 may be set as the vicinity gradation range. The neighborhood gradation range is appropriately set according to the purpose and may be set in any manner. In this embodiment, it is assumed that a value 70% of the total number of pixels of the input image (1920 × 1080 × 0.7 = 14551520) is obtained as the distribution amount S1.

  The gradation value correction unit 104 outputs the first gradation value P1 as it is when the distribution amount S1 is equal to or smaller than the predetermined threshold value, and the first gradation value P1 when the distribution amount S1 is larger than the predetermined threshold value. Is corrected and output. Details of the gradation value correction unit 104 will be described later.

The γ curve determination unit 105 determines the γ curve according to the gradation value output from the gradation value correction unit 104.
In this embodiment, the γ curve determination unit 105 stores a plurality of γ curves in association with the gradation values. Specifically, as shown in FIG. 5, 64 γ curves (γ curves of curve numbers B0 to B63) respectively corresponding to the gradation values 0 to 63 are stored. The γ curve determination unit 105 selects (determines) a γ curve corresponding to the gradation value (a γ curve having the same curve number as the gradation value) from the stored γ curves. Remembered γ
The curve is a γ curve that increases the gradation on the low gradation side as the curve number is small and decreases the gradation on the low gradation side as the curve number is large. The γ curve is a γ curve that enhances the gradation of the gradation range for the predetermined gradation including the corresponding gradation value (the corresponding gradation value and the gradation range in the vicinity thereof). A dotted line in FIG. 5 is a conversion straight line when the input signal is linearly converted into an output signal. The γ curve determined by the γ curve determination unit 105 is not limited to these. Any γ curve may be used as long as it is a γ curve corresponding to the gradation value (for example, a γ curve that enhances the gradation of the gradation range for the predetermined gradation including the corresponding gradation value).
The γ conversion unit 106 converts the gradation of the input image using the γ curve determined by the γ curve determination unit 105 (performs γ conversion processing). Then, an image (output image; output image signal) subjected to the γ conversion process is output.

(Details of gradation value correction unit 104)
Hereinafter, the gradation value correction unit 104 will be described in detail.
When the value of the distribution amount S1 is large, the first gradation value P1 may not be a gradation value of a flat area with a large area (typically a background), that is, a gradation value of an area to be noted. high. Therefore, in such a case, the gradation property of the γ curve (the gradation range (near gradation range) for the predetermined gradation including the first gradation value P1) is increased according to the first gradation value P1. It is not preferable to determine (γ curve). Specifically, in such a case, the gradation characteristics in the vicinity gradation range are low compared to an image in which gradation is converted using a γ curve determined according to the first gradation value P1. It is desirable to set a γ curve for obtaining an image. By setting such a γ curve, the difference in brightness between the area having the first gradation value P1 as the main gradation value and the other areas becomes large, and a good contrast feeling can be expressed.
On the other hand, when the value of the distribution amount S1 is small, the first gradation value P1 is likely to be a gradation value of a region to be noted (even if the value of the distribution amount S1 is small, the first gradation value P1). (Because P1 is a gradation value with concentrated frequencies). Therefore, in such a case, the γ curve may be determined according to the first gradation value P1. Thereby, the gradation (visibility) of the region to be noted can be improved.

  Therefore, in this embodiment, the gradation value correction unit 104 determines that the first gradation value P1 is the gradation value of the region to be noted when the distribution amount S1 is equal to or less than the predetermined threshold value, and One gradation value P1 is output as it is. When the distribution amount S1 is larger than the predetermined threshold value, it is determined that the first gradation value P1 is not the gradation value of the region to be noticed, the first gradation value P1 is corrected, and the second gradation value P2 is corrected. Output as. Specifically, the γ curve determined according to the corrected first gradation value is converted into an image in which the gradation is converted using the γ curve determined according to the first gradation value before correction. In comparison, the first gradation value is corrected so as to obtain a γ curve for obtaining an image with a low gradation in the vicinity gradation range. In the present embodiment, a value that is 50% of the total number of pixels of the input image (1920 × 1080 × 0.5 = 1036800) is set as the predetermined threshold value. However, the predetermined threshold is not limited to this value. The predetermined threshold may be appropriately set according to the purpose (for example, to what extent is the region to be noted), and may be set in any manner. The predetermined threshold value may be set according to the width of the neighborhood gradation range, for example.

  An example of a specific configuration of the gradation value correction unit 104 will be described with reference to FIG. FIG. 4 is a block diagram illustrating an example of a functional configuration of the gradation value correction unit 104.

  The offset processing unit 109 subtracts a predetermined offset value f (predetermined threshold value) from the distribution amount S1, and outputs the result as an offset distribution amount S2. Further, when the subtraction result is a negative value, the offset processing unit 109 outputs 0 as the offset distribution amount S2. In this embodiment, the offset value f (predetermined threshold value) is 50% of the total number of pixels of the input image (1920 × 1080 × 0.5 = 1036800), so that the value of the offset distribution amount S2 is ( S1-f = 1451520-1036800 =) 414720 is obtained.

  The gain processing unit 110 multiplies the offset distribution amount S2 by a predetermined gain value and outputs the result as a gain distribution amount S3. In this embodiment, a gain value is obtained by dividing a predetermined value (100 in this embodiment) by the total number of pixels of the input image. Therefore, in this embodiment, (414720 × 100 ÷ (1920 × 1080) =) 20 is obtained as the value of the gain distribution amount S3.

The addition processing unit 111 adds the gain distribution amount S3 to the first gradation value P1, and outputs the result as the second gradation value P2. The addition processing unit 111 outputs 63 as the second gradation value when the addition result is 64 or more. Therefore, in this embodiment, (32 + 20 =) 52 is obtained as the second gradation value P2. As a result, as shown by the solid line in FIG. 6, the gradation corresponding to the predetermined gradation including the gradation value 32 is compared with the γ curve of the curve number B52 (the gradation is converted using the γ curve of the curve number B32). (Γ curve for obtaining an image having a low gradation in the gradation range) is selected. As a result, as shown in FIG. 7, it is possible to reduce the gradation of the region B which is not a notable region (darken the region B), and the difference in brightness between the region A and the region B becomes large, which is good. It can express a sense of contrast.
Further, as the value of the distribution amount S1 is larger than the predetermined threshold value, the first gradation value P1 is more likely to be a gradation value in a region that should not be noted. In the present embodiment, as the value of the distribution amount S1 is larger than a predetermined threshold, a γ curve for obtaining an image having a lower gradation in the gradation range for the predetermined gradation including the first gradation value P1 is determined. The first gradation value P1 is corrected so as to be selected. Therefore, more appropriate γ conversion processing can be performed according to the value of the distribution amount S1.
Further, according to the present embodiment, since the value of 50% of the total number of pixels of the input image is set as the offset value f, the gain is obtained when the distribution amount S1 is equal to or less than the value of 50% of the total number of pixels of the input image. The value of the distribution amount S3 is 0. Therefore, in such a case, the first gradation value P1 is output as it is as the gradation value of the region to be noted. As a result, the gradation (visibility) of a region to be noted can be improved.

As described above, according to the present embodiment, the distribution amount S1 and the predetermined gradation value are used to determine whether the tone value (concentrated tone) where the frequency is concentrated is the tone value of the region to be noted or the background tone value. It is possible to make a determination by a simple method of comparing with the threshold value. As a result, γ conversion processing using an appropriate γ curve can be performed.
Specifically, when the concentrated gradation is a gradation value of a region to be noted, γ conversion processing is performed by a γ curve corresponding to the concentrated gradation, so that the gradation property (visual recognition) of the region to be noted is visually recognized. Property). Further, when the concentrated gradation is not the gradation value of the region to be noted, a predetermined value including the concentrated gradation is compared with an image in which the gradation is converted using a γ curve determined according to the concentrated gradation. A γ conversion process is performed using a γ curve for obtaining an image with a low gradation in the gradation range corresponding to the gradation. For this reason, the difference in brightness between the area having the concentrated gradation as the main gradation value and the other areas becomes large, and a good contrast feeling can be expressed.

Note that the correction method of the first gradation value P1 is not limited to the above method. Compared to an image whose gradation is converted using a γ curve determined according to the first gradation value P1 before correction, the gradation characteristics of a gradation range for a predetermined gradation including the first gradation value P1 The first gradation value P1 may be corrected so that a γ curve for obtaining an image with a low image quality is determined. For example, a predetermined value may be added to the first gradation value P1 to obtain a corrected first gradation value (the larger the distribution amount S1 is, the larger the value is added to the first gradation value P1). May be) Further, the offset processing and gain processing are not limited to the above methods. For example, processing may be performed so that the values of the offset distribution amount S2 and the gain distribution amount S3 become smaller as the distribution amount S1 is larger than a predetermined threshold. In that case, the second gradation value P2 may be increased as the value of the gain distribution amount S3 is decreased.
In this embodiment, the search range of the concentrated gradation is set to the range of gradation values 0 to 63, but the search range is not limited to this. For example, gradation values 0 to 127 may be used as the search range. In addition, the gradation range (low gradation region) on the low gradation side is not limited, and an intermediate gradation region such as gradation values 32 to 95 or a high gradation region (high gradation regions) such as gradation values 128 to 255 is used. The side gradation range) may be used as the search range.
In this embodiment, the case where the gradation histogram is a luminance histogram has been described. However, the gradation histogram may be a gradation histogram of RGB signals of an input image, or a histogram in another signal format. Also good.

<Example 2>
Hereinafter, an image processing apparatus and an image processing method according to Embodiment 2 of the present invention will be described. Note that description of functions and the like similar to those of the first embodiment is omitted. FIG. 8 is a block diagram illustrating a functional configuration of the image processing apparatus 200 according to the present embodiment. In FIG. 8, the same functions (components) as those in FIG.
The image processing apparatus 200 according to the present embodiment further includes a divided gradation histogram generation unit 201 in addition to the functions of the image processing apparatus 100 according to the first embodiment.

  The divided gradation histogram generation unit 201 divides the input image into a plurality of regions (divided regions), and generates a gradation histogram for each divided region. In this embodiment, as shown in FIG. 9, the input image is divided into 4 × 4 (regions 1 to 16) divided regions, and a gradation histogram is generated for each divided region. The number of divisions is not limited to 4 × 4. For example, the input image may be divided into 3 × 3 divided regions, or may be divided into divided regions such as 6 × 6, 3 × 4, and 4 × 3.

The concentrated gradation detecting unit 102 detects a gradation value (concentrated gradation) where the frequency is concentrated from each of the gradation histograms for each area, and selects one of the detected gradation values for each area on the first floor. Output as adjustment value P1.
In this embodiment, the concentrated gradation detection unit 102 has a gradation range on the low gradation side (low gradation area; for example, a range of gradation values 0 to 63) and a gradation range on the high gradation side (high gradation). A concentrated gradation is detected from each of the regions (e.g., a range of gradation values 128 to 255). The tone value on the lowest tone side among the tone values for each region detected from the low tone region, and the tone on the highest tone side among the tone values for each region detected from the high tone region The values are output as first gradation values P1-1 and P1-2, respectively.

The distribution amount calculation unit 103 calculates the sum of the frequencies in the gradation range (neighboring gradation range) for a predetermined gradation including the first gradation value P1 in the gradation histogram for one screen of the input image, Output as distribution amount S1.
In the present embodiment, the distribution amount calculation unit 103 calculates the sum of the frequencies within the gradation range for the predetermined gradation including the first gradation value P1-1, and outputs it as the distribution amount S1-1. Further, the sum of the frequencies within the gradation range for the predetermined gradation including the first gradation value P1-2 is calculated and output as the distribution amount S1-2.
Further, in this embodiment, when calculating the distribution amount S1, the gradation histogram for one screen generated by the gradation histogram generation unit 101 is used. Thereby, the calculation amount can be reduced as compared with the calculation using the gradation histogram for each region.

  Other functions are the same as those in the first embodiment. Specifically, the same processing as the first embodiment (processing by the tone value correcting unit 104) is performed for each of the first tone values P1-1 and P1-2. Then, a γ curve is determined according to the gradation value output from the gradation value correction unit 104, and γ conversion processing is performed on the image using the determined γ curve.

According to the present embodiment, a concentrated gradation is detected for each divided region, and any one of them is set as the first gradation value, so that the first gradation value is the gradation value of the region to be noted. It becomes possible to determine more accurately. For example, it is possible to prevent a gradation value whose frequency is not peaked in the partial area from being peaked by counting the entire image and being erroneously detected as a concentrated gradation. In addition, when the concentrated gradation is the gradation value of the region to be noticed, the γ conversion processing by the γ curve corresponding to the concentrated gradation is performed, so the gradation property (visibility) of the region to be noted is reduced. Can be increased. Further, when the concentrated gradation is not the gradation value of the region to be noted, a predetermined value including the concentrated gradation is compared with an image in which the gradation is converted using a γ curve determined according to the concentrated gradation. A γ conversion process is performed using a γ curve for obtaining an image with a low gradation in the gradation range corresponding to the gradation. For this reason, the difference in brightness between the area having the concentrated gradation as the main gradation value and the other areas becomes large, and a good contrast feeling can be expressed.

  In this embodiment, both the lowest gradation side concentrated gradation and the highest gradation side concentrated gradation are set as the first gradation value, but the first gradation value is not limited to this. Either the lowest gradation side concentrated gradation or the highest gradation side concentrated gradation may be set as the first gradation value. If at least the lowest gradation side concentrated gradation is the first gradation value, blackout can be suppressed when the first gradation value is a gradation value of a region to be noted. Further, if at least the concentrated gradation on the highest gradation side is the first gradation value, overexposure can be suppressed when the first gradation value is a gradation value of a region to be noted.

<Other examples>
The present invention can be applied to a case where each block of the image processing apparatus is implemented by hardware or a software process using a computer, and similar effects can be obtained. When the computer executes the supplied program code, the software program code itself realizes the functions described in the above embodiments. The program code itself and means for supplying the program code to a computer, for example, a storage medium storing the program code constitute the present invention. As a storage medium for storing the program code, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

The present invention is also applicable to the case where the functions described in the above embodiments are realized in cooperation with an OS (operating system) or other application software running on the computer. Can do. Even in such a case, such a program code is included in the embodiment of the present invention.
The supplied program code may be stored in a memory provided in a function expansion board of a computer or a function expansion unit connected to the computer. The CPU or the like provided in the function expansion board or function expansion unit performs part or all of the actual processing based on the instruction of the program code, and the functions described in the above embodiments are realized by the processing. Also good. Even in such a case, such a program code is included in the embodiment of the present invention.

DESCRIPTION OF SYMBOLS 100 Image processing apparatus 101 Gradation histogram generation part 102 Concentration gradation detection part 103 Distribution amount calculation part 104 Gradation value correction part 105 γ curve determination part 106 γ conversion part

Claims (9)

Generating means for generating a gradation histogram from the input image;
Detecting means for detecting a gradation value at which the frequency is concentrated from the gradation histogram and outputting it as a first gradation value;
Calculating means for calculating the sum of the frequencies in the vicinity gradation range which is a gradation range for the predetermined gradation including the first gradation value in the gradation histogram, and outputting as a distribution amount;
A correction unit that outputs the first gradation value as it is when the distribution amount is equal to or smaller than a predetermined threshold value, and corrects and outputs the first gradation value when the distribution amount is larger than the predetermined threshold value. When,
Determining means for determining a γ curve according to the gradation value output by the correcting means;
Γ conversion means for converting the gradation of the image using the determined γ curve;
Have
The γ curve determined according to the first gradation value after correction is compared with the neighborhood floor compared to the image whose gradation is converted using the γ curve determined according to the first gradation value before correction. An image processing apparatus having a γ curve for obtaining an image having a low gradation property in a gradation range. The correction means corrects the first gradation value so that a γ curve for obtaining an image having a lower gradation property in the neighboring gradation range is determined as the distribution amount is larger than a predetermined threshold value. The image processing apparatus according to claim 1, wherein:   When the distribution amount is larger than the predetermined threshold, the correction unit subtracts the predetermined threshold from the distribution amount, multiplies a value obtained by the subtraction by a predetermined gain value, and performs the multiplication. The image processing apparatus according to claim 1, wherein the obtained value is added to the first gradation value and output as the corrected first gradation value. The generation unit divides the image into a plurality of regions, generates a gradation histogram for each region,
The detecting means detects a gradation value in which the frequency is concentrated from each of the gradation histograms for each region, and outputs any of the gradation values detected for each region as the first gradation value,
The calculation means calculates a total sum of frequencies in the neighboring gradation range in a gradation histogram for one screen of the image and outputs it as a distribution amount. The image processing apparatus according to item. The image processing according to claim 4, wherein the detection unit outputs a gradation value on the lowest gradation side among gradation values detected for each region as the first gradation value. apparatus. The image processing apparatus according to claim 4, wherein the detection unit outputs a gradation value on a highest gradation side among gradation values detected for each region as the first gradation value. .   The image processing apparatus according to claim 1, wherein the gradation histogram is a gradation histogram of a luminance signal of the input image.   The image processing apparatus according to claim 1, wherein the gradation histogram is a gradation histogram of RGB signals of the input image. A generation step for generating a gradation histogram from the input image;
A detection step of detecting a gradation value in which the frequency is concentrated from the gradation histogram and outputting it as a first gradation value;
A calculation step of calculating a sum of the frequencies in a neighboring gradation range that is a gradation range for a predetermined gradation including the first gradation value in the gradation histogram, and outputting as a distribution amount;
A correction step of outputting the first gradation value as it is when the distribution amount is equal to or less than a predetermined threshold, and correcting and outputting the first gradation value when the distribution amount is larger than the predetermined threshold. When,
A determination step of determining a γ curve according to the gradation value output in the correction step;
A γ conversion step of converting the gradation of the image using the determined γ curve;
Have
The γ curve determined according to the first gradation value after correction is compared with the neighborhood floor compared to the image whose gradation is converted using the γ curve determined according to the first gradation value before correction. An image processing method, wherein the image processing method is a γ curve for obtaining an image with a low gradation in a gradation range.

Images