JP2016045392A - Image processor, image processing method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique capable of acquiring image data for display whose impression is close to an image represented by base image data and in which a dynamic range is appropriately expanded.SOLUTION: An image processor of this invention includes: first acquisition means for acquiring second image data; second acquisition means for acquiring first correspondence data representing a correspondence relation between a first gradation value being a gradation value which first image data can take and a second gradation value being a gradation value which the second image data can take; determination means for determining a portion of the range of the second gradation value as a maintenance range; generation means for generating second correspondence data representing a correspondence relation between the second gradation value and a gradation value for display being a gradation value which image data for display can take on the basis of the first correspondence data, the maintenance range, and a luminance range being the range of luminance which an image displayed on the screen of a display part can take; and conversion means for converting the second image data into image data for display by using the second correspondence data.SELECTED DRAWING: Figure 1

Description

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

  In recent years, HDR (High Dynamic Range) display has been performed in which a multi-bit image data is used to obtain a display image having a wider dynamic range than a conventional display image (image displayed on the screen).

  A method of recording HDR image data (multi-bit image data) having a wide dynamic range by dividing the HDR image data into base image data and gradation difference data, and recording the base image data and gradation difference data Has been proposed (Patent Document 1). That is, a format using base image data and gradation difference data has been proposed as one of data formats of HDR image data.

The base image data is small-bit image data obtained by compressing the dynamic range of HDR image data (range compression). The base image data is image data of about 8 bit gradation, for example. The producer of the image data sets the dynamic range of the base image data so that an important gradation range in the image data is included in the dynamic range of the base image data among the wide dynamic range (gradation range) of the HDR image data. decide.
The luminance difference data is data representing a correspondence relationship between the gradation value (for example, luminance value) of the base image data and the gradation value of the HDR image data.
By converting the gradation value of the base image data into the gradation value of the HDR image data using the luminance difference data, the HDR image data can be obtained (restored).

The related art relating to the gradation conversion processing for converting the gradation value is also disclosed in Patent Document 2.
In the technique disclosed in Patent Document 2, the distribution of gradation values of target image data (image data to be subjected to gradation conversion processing) is analyzed, and gradation conversion processing based on the analysis result is performed on the target image data.

  The range of brightness that the display image can take is wider than the range of brightness that corresponds to the range of tone values that the base image data can take, and the range of brightness that corresponds to the range of tone values that the HDR image data can take May be narrower. In this case, it is necessary to compress the dynamic range of the HDR image data so that the luminance range corresponding to the range of gradation values that the image data can take falls within the luminance range that the display image can take.

  However, when the technique of Patent Document 2 is used, the image data for display is obtained by subjecting the HDR image data to gradation conversion processing based on the analysis result of the gradation distribution (gradation value distribution) of the HDR image data. Is generated. The display image data is image data obtained by compressing the dynamic range of the HDR image data. The solid line in FIG. 14 indicates the conversion characteristic (first conversion characteristic) when the HDR image data is converted into the base image data, and the broken line in FIG. 14 indicates that the HDR image data is displayed using the technique of Patent Document 2. The conversion characteristic (2nd conversion characteristic) at the time of converting into data is shown. As shown in FIG. 14, characteristics that are significantly different from the first conversion characteristics may be set as the second conversion characteristics. For this reason, display image data representing an image whose impression is significantly different from the image represented by the base image data is obtained, and a display image having an impression significantly different from the impression intended by the producer may be obtained.

JP 2011-193511 A JP-A-3-126377

  An object of the present invention is to provide a technique capable of obtaining display image data having an impression close to that of an image represented by base image data and having a suitably expanded dynamic range.

The first aspect of the present invention is:
First acquisition means for acquiring second image data with an expanded dynamic range of the first image data;
First correspondence data representing a correspondence relationship between a first gradation value that is a gradation value that can be taken by the first image data and a second gradation value that is a gradation value that can be taken by the second image data. A second acquisition means for acquiring;
Determining means for determining a partial range of the second gradation value range as a maintenance range;
Based on the first correspondence data, the maintenance range, and a luminance range that can be taken by the image displayed on the screen of the display unit, the second gradation value and the display image data are taken. Generating means for generating second correspondence data representing a correspondence relationship between display gradation values that are obtained gradation values;
Conversion means for converting the second image data into the display image data using the second correspondence data;
Have
The luminance range is wider than the luminance range corresponding to the first gradation value range and narrower than the luminance range corresponding to the second gradation value range;
The generation unit converts the second gradation value belonging to the maintenance range into the same value as the first gradation value corresponding to the second gradation value, and has a dynamic range larger than that of the first image data. The image processing apparatus is characterized in that the second correspondence data is generated so that the wide display image data is generated.

The second aspect of the present invention is:
A first acquisition step of acquiring second image data with an expanded dynamic range of the first image data;
First correspondence data representing a correspondence relationship between a first gradation value that is a gradation value that can be taken by the first image data and a second gradation value that is a gradation value that can be taken by the second image data. A second obtaining step to obtain;
A determination step of determining a partial range of the second gradation value range as a maintenance range;
Based on the first correspondence data, the maintenance range, and a luminance range that can be taken by the image displayed on the screen of the display unit, the second gradation value and the display image data are taken. A generation step of generating second correspondence data representing a correspondence relationship with the display gradation value which is a gradation value to be obtained;
A conversion step of converting the second image data into the display image data using the second correspondence data;
Have
The luminance range is wider than the luminance range corresponding to the first gradation value range and narrower than the luminance range corresponding to the second gradation value range;
In the generating step, the second gradation value belonging to the maintenance range is converted to the same value as the first gradation value corresponding to the second gradation value, and the dynamic range is greater than that of the first image data. In the image processing method, the second correspondence data is generated so that the wide display image data is generated.

  A third aspect of the present invention is a program that causes a computer to execute each step of the above-described image processing method.

  According to the present invention, it is possible to obtain display image data having an impression close to that of the image represented by the base image data and having a suitably expanded dynamic range.

1 is a block diagram illustrating an example of a functional configuration of an image processing apparatus according to a first embodiment. The figure which shows an example of the tone map which concerns on Example 1, and the tone map for panel conversion The figure which shows an example of the look-up table showing the coefficient which concerns on Example 1. FIG. FIG. 10 is a diagram illustrating an example of the operation of the image processing apparatus according to the first embodiment. The figure which shows an example of the look-up table showing the coefficient which concerns on Example 1. FIG. 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 the tone map which concerns on Example 2, and the tone map for panel conversion The figure which shows an example of the look-up table showing the coefficient which concerns on Example 2. FIG. The figure which shows an example of the look-up table showing the coefficient which concerns on Example 2. FIG. FIG. 9 is a block diagram illustrating an example of a functional configuration of an image processing apparatus according to a third embodiment. The figure which shows an example of the ratio data which concern on Example 3, and the ratio data for panel conversion The figure which shows an example of the look-up table showing the coefficient which concerns on Example 3. FIG. FIG. 10 is a diagram illustrating an example of the operation of the image processing apparatus according to the third embodiment. The figure which shows an example of operation | movement of the conventional image processing apparatus.

<Example 1>
Hereinafter, an image processing apparatus and an image processing method according to Embodiment 1 of the present invention will be described.
FIG. 1 is a block diagram illustrating an example of a functional configuration of the image processing apparatus 100 according to the present embodiment.
The image processing apparatus 100 includes a decoding processing unit 102, a tone map generation unit 103, a maintenance range determination unit 104, a tone map conversion unit 105, a range compression processing unit 106, a display unit 107, and the like.
Note that the image processing apparatus 100 may not include the display unit 107. The display unit 107 may be an image display device that is separate from the image processing device 100.

An HDR file 101 is input to the image processing apparatus 100.
The HDR file 101 is a data file including first image data and corresponding data (gradation difference data).
The first image data is image data in which the dynamic range of the second image data is compressed (range compression). In other words, the second image data is image data obtained by expanding the dynamic range of the first image data. The first image data is, for example, image data of about 8 bit gradation. In this embodiment, the first image data is described as “base image data”, and the second image data is described as “HDR image data”.
The correspondence data represents a correspondence relationship between a first gradation value that is a gradation value that can be taken by the first image data and a second gradation value that is a gradation value that can be taken by the second image data. In the present embodiment, a lookup table (reverse tone map R_TM) that represents the first gradation value as an input value and the second gradation value as an output value is used as the corresponding data. The gradation value is a pixel value, a luminance value, a color component value constituting the pixel value, or the like. The color component values are R value, G value, B value, and the like.
Note that the first image data and the corresponding data may be acquired separately.

  The decode processing unit 102 generates (restores; acquires) HDR image data using the base image data and the reverse tone map R_TM included in the HDR file 101 (first acquisition process). Then, the decode processing unit 102 outputs the generated HDR image data.

The tone map generation unit 103 generates (acquires) the tone map TM_A based on the reverse tone map R_TM included in the HDR file 101 (second acquisition process). Then, the tone map generating unit 103 outputs the generated tone map TM_A.
The tone map TM_A is first correspondence data representing the correspondence between the first gradation value and the second gradation value. Specifically, the tone map TM_A is a lookup table (first lookup table) that represents the second gradation value as an input value and represents the first gradation value as an output value. That is, the tone map TM_A is a lookup table in which the input value and the output value of the reverse tone map R_TM are interchanged.

  The maintenance range determination unit 104 determines a partial range of the second gradation value range as the maintenance range. Then, the maintenance range determination unit 104 outputs maintenance range information representing the determined maintenance range. In the present embodiment, the maintenance range determination unit 104 determines one second gradation value as the maintenance gradation value. The sustain gradation value can also be said to be “a maintenance range including one second gradation value”.

The tone map conversion unit 105 generates a panel conversion tone map P_TM_a based on the tone map TM_A, the maintenance range (maintenance gradation value), and the luminance range (panel luminance range). In this embodiment, a panel conversion tone map P_TM_a is generated by correcting each first gradation value represented by the tone map TM_A. That is, by correcting each input value of the tone map TM_A, a panel conversion tone map P_TM_a is generated. Then, the tone map conversion unit 105 outputs the generated panel conversion tone map P_TM_a.
The panel luminance range is a luminance range that an image (display image) displayed on the screen of the display unit 107 can take. In this embodiment, it is assumed that the panel luminance range is wider than the luminance range corresponding to the first gradation value range and narrower than the luminance range corresponding to the second gradation value range.
The panel conversion tone map P_TM_a is second correspondence data representing a correspondence relationship between the second gradation value and the display gradation value that is a gradation value that can be taken by the display image data. Specifically, the panel conversion tone map P_TM_a is a lookup table (second lookup table) that represents the second gradation value as an input value and represents the display gradation value as an output value.

In the present embodiment, the panel conversion tone map P_TM_a is generated so that the following two processes are performed by the range compression processing unit 106.
A process of converting the second gradation value equal to the maintenance gradation value to the same value as the first gradation value corresponding to the second gradation value. Display image data having a wider dynamic range than the base image data. Processing to be generated By performing the two processes described above, the dynamic range of the HDR image data is compressed so that the luminance range corresponding to the range of gradation values that the image data can take is within the panel luminance range. be able to. Further, it is possible to obtain display image data having an impression close to the image represented by the base image data and having a suitably expanded dynamic range.
The second process of the two processes described above is, for example, the following two processes.
A value that is greater than the first gradation value corresponding to the second gradation value and that is equal to or less than the gradation value corresponding to the maximum luminance in the panel luminance range. The second gradation value smaller than the sustain gradation value is smaller than the first gradation value corresponding to the second gradation value and the gradation corresponding to the minimum luminance in the panel luminance range. Processing to convert to a value greater than or equal to

Information indicating the panel luminance range may be recorded in advance in the image processing apparatus 100 or may be acquired from the outside of the image processing apparatus 100. For example, when the display unit 107 is an image display device that is separate from the image processing device 100, information indicating the panel luminance range may be acquired from the display unit 107.

The range compression processing unit 106 uses the panel conversion tone map P_TM_a generated by the tone map conversion unit 105 to convert the HDR image data generated by the decoding processing unit 102 into display image data. Then, the range compression processing unit 106 outputs display image data. In this embodiment, display image data is generated by compressing the dynamic range of the HDR image data so that the luminance range corresponding to the range of gradation values that can be taken by the image data matches the panel luminance range. .
Note that the luminance range corresponding to the range of gradation values that the display image data can take may be narrower than the panel luminance range.

  The display unit 107 displays the image represented by the display image data output from the range compression processing unit 106 on the screen.

In this embodiment, the contrast ratio is 2000: 1, the minimum luminance of the panel luminance range is 0.1 [cd / m ² ], and the maximum luminance of the panel luminance range is 200 [cd / m ² ]. Assume that the display unit 107 having the capability is used.
In this embodiment, the maximum value BI_H of the gradation values (first gradation value BI) that can be taken by the base image data is 1, and the luminance corresponding to the maximum value BI_H is 100 [cd / m ² ]. And In this embodiment, the maximum value BI_H is called “100% white”, and the luminance corresponding to 100% white is called “standard luminance”.
In the present embodiment, it is assumed that the minimum value BI_L of the first gradation value BI is 0.002 and the luminance corresponding to the minimum value BI_L is 0.2 [cd / m ² ]. In this embodiment, the minimum value BI_L is referred to as “0.2% black”.
In this embodiment, the maximum value HI_H of the gradation values (second gradation value HI) that can be taken by the HDR image data is 10, and the luminance corresponding to the maximum value HI_H is 1000 [cd / m ² ]. And
In this embodiment, it is assumed that the minimum value HI_L of the second gradation value HI is 0.0001, and the luminance corresponding to the minimum value HI_L is 0.01 [cd / m ² ].

Therefore, the maximum value PI_H of the display gradation value PI is 2.0 (= (100% white gradation value) × (maximum luminance in the panel luminance range) / (standard luminance)).
The minimum value PI_L of the display gradation value PI is 0.001 (= (100% white gradation value) × (minimum luminance in the panel luminance range) / (standard luminance)).

In the present embodiment, the range compression processing unit 106 compresses the range of gradation values that the image data can take from a range of 0.0001 or more and 10 or less to a range of 0.001 or more and 2 or less. The HDR image data is converted into display image data.
A pixel having a gradation value larger than 100% white (1) is displayed with a luminance higher than the standard luminance. In this embodiment, a pixel having a gradation value larger than 100% white is called an “over white pixel”.
Pixels whose gradation value is smaller than 0.2% black (0.002) are displayed with a luminance lower than the luminance corresponding to 0.2% black. In this embodiment, a pixel having a gradation value smaller than 0.2% black is referred to as an “under black pixel”.

FIG. 2 shows an example of the tone map TM_A and the panel conversion tone map P_TM_a. The solid line in FIG. 2 indicates the tone map TM_A, and the broken line in FIG. 2 indicates the panel conversion tone map P_TM_a. The horizontal axis in FIG. 2 indicates an input value (second gradation value), and the vertical axis in FIG. 2 indicates an output value (first gradation value and display gradation value).
FIG. 2 shows a tone map TM_A for reducing the gradation of the second gradation value range (bright part) on the high gradation side and the second gradation value range (dark part) on the low gradation side. Yes. Specifically, the tone map TM_A in which more first gradation values than the bright and dark portions are assigned to the range of the second gradation value in the intermediate portion is shown.

  The maintenance range determination unit 104, tone map conversion unit 105, and range compression processing unit 106 will be described in more detail with reference to FIGS.

  The maintenance range determination unit 104 determines one first gradation value BI as a base gradation value using the reverse tone map R_TM (or tone map TM_A), and a second gradation value corresponding to the base gradation value HI is detected. Then, the maintenance range determination unit 104 outputs the detected second gradation value HI as the maintenance gradation value ST_P1. In this embodiment, an intermediate value of the first gradation value BI is detected as the base gradation value. As described above, in the present embodiment, the range of the first gradation value BI is 0.002 or more and 1 or less. Therefore, 0.5 (specifically, 0.499) is detected as the intermediate value of the first gradation value BI. In the tone map TM_A of FIG. 2, the second gradation value HI = 0.1 is associated with the first gradation value BI = 0.5. Therefore, 0.1 is detected as the second gradation value HI corresponding to the intermediate value of the first gradation value BI.

The base gradation value is not limited to the intermediate value of the first gradation value BI. The base gradation value may be a fixed value determined in advance by the manufacturer, or may be a value set by the user. The base gradation value may be determined according to the installation environment of the image processing apparatus 100 and the display unit 107 (image display apparatus).
Note that the method for determining the sustain gradation value is not limited to the above method. For example, the second gradation value corresponding to the intermediate value of the gradation values of the base image data may be determined as the maintenance gradation value. The sustain gradation value may be determined according to the installation environment of the image processing apparatus 100 and the display unit 107 (image display apparatus). The sustain gradation value may be determined according to a user operation.

The tone map conversion unit 105 generates a lookup table LUT_A based on the tone map TM_A, the sustain gradation value, and the panel luminance range. The lookup table LUT_A represents the coefficient C of the first gradation value BI corresponding to the second gradation value HI for each second gradation value HI.
The tone map conversion unit 105 multiplies the first gradation value by the coefficient of the first gradation value for each first gradation value represented by the tone map TM_A, thereby obtaining the panel conversion tone map P_TM_a. Generate. That is, the panel conversion tone map P_TM_a is generated by converting each first gradation value BI represented by the tone map TM_A into the display gradation value PI using the following Expression 1. In Equation 1, P_TM_a (x) is the display gradation value PI corresponding to the second gradation value x, and TM_A (x) is the first gradation value BI corresponding to the second gradation value x. Yes, LUT_A (x) is a coefficient C corresponding to the second gradation value x.

P_TM_a (x) = TM_A (x) × LUT_A (x) (Formula 1)

In this embodiment, 1 is set as the coefficient C of the first gradation value BI corresponding to the second gradation value HI belonging to the maintenance range.
Specifically, 1 is set as the coefficient C of the first gradation value BI corresponding to the sustain gradation value.

In this embodiment, a value larger than 1 is set as the coefficient C of the first gradation value BI corresponding to the second gradation value HI that is larger than the maximum gradation value in the maintenance range. Then, a value smaller than 1 is set as the coefficient C of the first gradation value BI corresponding to the second gradation value HI smaller than the minimum gradation value of the maintenance range.

  Specifically, a value larger than 1 is set as the coefficient C of the first gradation value BI corresponding to the second gradation value HI that is larger than the sustain gradation value, and the second floor is smaller than the sustain gradation value. A value smaller than 1 is set as the coefficient C of the first gradation value BI corresponding to the tone value HI.

  More specifically, as the coefficient C of the first gradation value BI corresponding to the maximum value of the second gradation value HI, the gradation value corresponding to the maximum luminance of the panel luminance range with respect to the first gradation value BI. A percentage is set. As the coefficient C of the first gradation value BI corresponding to the minimum value of the second gradation value HI, the ratio of the gradation value corresponding to the minimum luminance of the panel luminance range to the first gradation value BI is set. Then, the coefficient C of the first gradation value BI corresponding to the second gradation value HI (maintenance gradation value) belonging to the maintenance range, and the first gradation value BI corresponding to the maximum value of the second gradation value HI. Interpolation processing using the coefficient C and the coefficient C of the first gradation value BI corresponding to the minimum value of the second gradation value HI is performed. Accordingly, the first gradation that does not correspond to the second gradation value HI belonging to the maintenance range, does not correspond to the maximum value of the second gradation value HI, and does not correspond to the minimum value of the second gradation value HI. The coefficient of value BI is determined.

  “The ratio of the gradation value corresponding to the maximum luminance of the panel luminance range to the first gradation value BI” may be referred to as “the ratio of the maximum luminance of the panel luminance range to the luminance corresponding to the first gradation value BI”. it can. The “ratio of the gradation value corresponding to the minimum luminance of the panel luminance range to the first gradation value BI” may be referred to as “the ratio of the minimum luminance of the panel luminance range to the luminance corresponding to the first gradation value BI”. it can.

In the present embodiment, the maximum luminance of the panel luminance range is 200 [cd / m ² ], which is twice the luminance (100 [cd / m ² ]) corresponding to the maximum value of the first gradation value BI. . The minimum luminance of the panel luminance range is 0.1 [cd / m ² ], and the luminance (0.2 [cd / m ² ]) corresponding to the minimum value of the first gradation value BI is 0.5. Is double.
Therefore, as shown in FIG. 3, 2 is set as the coefficient C of the first gradation value BI (maximum value) corresponding to the second gradation value HI = 10, and the second gradation value HI = 0.0001. 0.5 is set as the coefficient C of the first gradation value BI (minimum value) corresponding to. Further, 1 is set as the coefficient C of the first gradation value BI corresponding to the second gradation value HI = 0.1 (maintenance gradation value). Then, another coefficient C is determined by linear interpolation. Specifically, the other coefficient C is determined so that the coefficient C increases linearly as the second gradation value HI increases.

  The range compression processing unit 106 inputs the gradation value of the HDR image data to the panel conversion tone map P_TM_a, thereby converting the gradation value of the HDR image data into the gradation value of the display image data. Thereby, the HDR image data is converted into display image data. Specifically, when the gradation value of the HDR image data is xr, P_TM_a (xr) is obtained as the gradation value of the display image data.

When the look-up table of FIG. 3 is used, the tone map TM_A of FIG. 2 is converted into the panel conversion tone map P_TM_a of FIG. When the panel conversion tone map P_TM_a in FIG. 2 is used, the gradation value of the HDR image data is converted into the gradation value of the display image data as described below.
As the gradation value of the display image data, a larger gradation value is obtained as the gradation value of the HDR image data is larger.
A gradation value equal to the sustain gradation value is converted to the same value as the first gradation value corresponding to the gradation value.
A gradation value larger than the sustain gradation value is converted into a gradation value larger than the first gradation value corresponding to the gradation value.
A gradation value smaller than the sustain gradation value is converted into a gradation value smaller than the first gradation value corresponding to the gradation value.
A gradation value equal to the maximum value of the second gradation value is converted into a gradation value corresponding to the maximum luminance of the panel luminance range.
A gradation value equal to the minimum value of the second gradation value is converted into a gradation value corresponding to the minimum luminance in the panel luminance range.

  Thereby, it is possible to obtain a display image in which the appearance in the region where the gradation value of the HDR image data is equal to the sustain gradation value is equal to the image (base image) represented by the base image data. In addition, a display image having a wider dynamic range than the base image can be obtained. Specifically, an area where the gradation value of HDR image data is larger than the sustain gradation value is brighter than the base image, and an area where the gradation value of HDR image data is smaller than the maintenance gradation value is greater than the base image. Even dark display images can be obtained.

A specific example of processing of the image processing apparatus according to the present embodiment will be described with reference to FIG.
FIG. 4 shows a tone map TM_A, a panel change tone map P_TM_a, and each image. In the upper part of the right side of FIG. 4, an HDR image a that is an example of an image (HDR image) represented by the HDR image data is shown. In the middle part on the right side of FIG. 4, a base image b which is an example of a base image is shown. A display image c, which is an example of an image (display image) represented by the display image data, is shown in the lower part on the right side of FIG. The numerical value described in each image is a gradation value in an area where the numerical value is described.

  In the present embodiment, the base image b is converted into the HDR image a using an inverse tone map in which the input value and output value of the tone map TM_A are interchanged. Then, the tone map TM_A is converted into a panel change tone map P_TM_a. Thereafter, the HDR image a is converted into the display image c using the panel change tone map P_TM_a.

  In the tone map TM_A of FIG. 4, the second gradation value 2 is set to the first gradation value 0.9, 0.1 is set to the first gradation value 0.5, and the second gradation value is set to 0.005. A gradation value of 0.001 is associated. Therefore, when the inverse tone map in which the input value and the output value of the tone map TM_A in FIG. 4 are exchanged is used, the gradation value 0.9 of the left square of the base image b is converted to the gradation value 2. The gradation value 0.005 in the right square is converted into a gradation value 0.001, and the background gradation value 0.5 is converted into a gradation value 0.1. Thereby, the base image b is converted into the HDR image a.

In the panel change tone map P_TM_a in FIG. 4, the second gradation value 2 is associated with a display gradation value 1.8 greater than the first gradation value 0.9. In addition, the second gradation value is 0.1.
The display gradation value 0.5 that is the same as the first gradation value is associated with the second gradation value 0.001, and the display gradation value 0. 0 smaller than the first gradation value 0.005. 0015 is associated. Therefore, when the panel conversion tone map P_TM_a in FIG. 4 is used, the gradation value 2 in the left square of the HDR image a is converted to a gradation value 1.8 that is brighter than the base image. Then, the gradation value 0.001 of the right square is converted to a gradation value 0.0015 that is darker than the base image, and the background gradation value 0.1 is converted to the same gradation value 0.5 as that of the base image. Is done. Thereby, the HDR image a is converted into the display image c. As a result, it is possible to obtain a display image in which the appearance of the background is equal to the base image, the left square is brighter than the base image, and the right square is darker than the base image.

  As described above, according to this embodiment, the second gradation value equal to the sustain gradation value is converted into the same value as the first gradation value corresponding to the second gradation value, and the base image Display image data having a wider dynamic range than the data is generated. Thereby, it is possible to obtain display image data having an impression close to the image represented by the base image data and having a suitably expanded dynamic range.

  Note that the image data according to the present embodiment may be still image data (image data representing a still image) or moving image data (image data representing a moving image). For example, an inverse tone map of the scene may be acquired for each scene of the base image data that is moving image data. In that case, for each scene of the moving image data, a panel conversion tone map of the scene may be generated from the reverse tone map of the scene. Then, for each scene of moving image data, display image data for the scene may be generated using the panel conversion tone map of the scene.

It should be noted that the specific numerical values described in this embodiment are merely examples, and are not intended to limit the scope of the present invention. For example, the display unit 107 has a contrast ratio of 6000: 1, a minimum luminance of the panel luminance range of 0.05 [cd / m ² ], and a maximum luminance of the panel luminance range of 300 [cd / m ² ]. It may have a certain display capability. In this case, 3 is set as the coefficient C of the first gradation value BI (maximum value) corresponding to the second gradation value HI = 10, and the first floor corresponding to the second gradation value HI = 0.0001. 0.25 may be set as the coefficient C of the adjustment value BI (minimum value). Then, the HDR image data is converted into the display image data so that the range of gradation values that the image data can take is compressed from the range of 0.0001 or more and 10 or less to the range of 0.0005 or more and 3 or less. It only needs to be converted.

Note that the characteristics of the lookup table representing the coefficient C are not limited to the characteristics shown in FIG. For example, nonlinear interpolation may be performed to generate a look-up table having S-curve characteristics as shown in FIG.
The method for generating the panel conversion tone map is not limited to the above method. For example, a conversion function for converting a tone map into a panel conversion tone map may be generated based on the tone map, the maintenance range, and the panel luminance range. Then, without determining the coefficient C, the tone map may be converted to a panel conversion tone map using a conversion function.
Note that the maximum value of the second gradation value may be converted into a gradation value corresponding to a luminance lower than the maximum luminance of the panel luminance range. The minimum value of the second gradation value may be converted into a gradation value corresponding to a luminance higher than the minimum luminance of the panel luminance range.
In the present embodiment, the example in which the base image data and the reverse tone map are acquired from the outside has been described, but the present invention is not limited to this. For example, HDR image data and a tone map (or reverse tone map) may be acquired from the outside.

<Example 2>
Hereinafter, an image processing apparatus and an image processing method according to Embodiment 2 of the present invention will be described.
In the first embodiment, the example in which the maintenance gradation value (the maintenance range including one second gradation value) is determined as the maintenance range has been described.
In this embodiment, an example in which a maintenance range including a plurality of second gradation values is determined will be described.

FIG. 6 is a block diagram illustrating an example of a functional configuration of the image processing apparatus 200 according to the present embodiment.
The image processing apparatus 200 includes a decoding processing unit 102, a tone map generation unit 103, a maintenance range determination unit 201, a tone map conversion unit 202, a range compression processing unit 106, a display unit 107, and the like.
In FIG. 6, the same functional parts as those in the first embodiment (FIG. 1) are denoted by the same reference numerals as those in FIG. In this embodiment, the tone map output from the tone map generation unit 103 is described as “tone map TM_B”.
In this embodiment, it is assumed that the panel luminance range, the dynamic range of base image data, the dynamic range of HDR image data, and the dynamic range of display image data are the same as those in the first embodiment.

  The maintenance range determination unit 201 has the same function as the maintenance range determination unit 104 of the first embodiment. That is, the maintenance range determination unit 201 determines a partial range of the second gradation value range as the maintenance range. And the maintenance range determination part 201 outputs the maintenance range information showing the determined maintenance range. However, the maintenance range determination unit 201 determines two second gradation values as the maximum gradation value and the minimum gradation value of the maintenance gradation values.

The tone map conversion unit 202 has the same function as the tone map conversion unit 105 of the first embodiment. That is, the tone map conversion unit 202 generates a panel conversion tone map P_TM_b based on the tone map TM_B, the maintenance range, and the panel luminance range. Then, the tone map conversion unit 202 outputs the generated panel conversion tone map P_TM_b.
However, in this embodiment, the panel conversion tone map P_TM_b is generated so that the following two processes are performed in the range compression processing unit 106.
A process of converting the second gradation value belonging to the maintenance range into the same value as the first gradation value corresponding to the second gradation value. Display image data having a wider dynamic range than the base image data is generated. Processing By performing the two processes described above, the dynamic range of the HDR image data can be compressed so that the luminance range corresponding to the range of gradation values that the image data can take is within the panel luminance range. it can. Further, it is possible to obtain display image data having an impression close to the image represented by the base image data and having a suitably expanded dynamic range.
The second process of the two processes described above is, for example, the following two processes.
A gradation value that is larger than the first gradation value corresponding to the second gradation value and that is larger than the maximum gradation value in the maintenance range and that corresponds to the maximum luminance in the panel luminance range Processing for converting into the following values: The second gradation value smaller than the minimum gradation value of the maintenance range is smaller than the first gradation value corresponding to the second gradation value and the minimum of the panel luminance range Processing to convert to a value greater than the gradation value corresponding to the brightness

FIG. 7 shows an example of the tone map TM_B and the panel conversion tone map P_TM_b. The solid line in FIG. 7 indicates the tone map TM_B, and the broken line in FIG. 7 indicates the panel conversion tone map P_TM_b. The horizontal axis in FIG. 7 indicates an input value (second gradation value), and the vertical axis in FIG. 7 indicates an output value (first gradation value and display gradation value).
FIG. 7 shows a tone map TM_B for reducing the gradation of the second gradation value range (bright part) on the high gradation side and the second gradation value range (dark part) on the low gradation side. Yes.

  The maintenance range determination unit 201 and the tone map conversion unit 202 will be described in more detail with reference to FIGS.

The maintenance range determination unit 201 uses the inverse tone map R_TM (or tone map TM_B) to determine the maximum gradation value and the minimum gradation value of the maintenance range.
Specifically, the maintenance range determination unit 201 determines the range (base range) of the first gradation value BI that is desired to maintain the appearance of the base image. Then, the maintenance range determination unit 201 determines the second gradation value corresponding to the minimum gradation value of the base range as the minimum gradation value ST_P2 of the maintenance range. In addition, the maintenance range determination unit 201 determines the second gradation value corresponding to the maximum gradation value of the base range as the maximum gradation value ST_P3 of the maintenance range. The minimum gradation value of the base range is a first gradation value equal to the first threshold value, and the maximum gradation value of the base range is a first gradation value equal to a second threshold value that is larger than the first threshold value.
Then, the maintenance range determination unit 201 outputs the minimum gradation value ST_P2 and the maximum gradation value ST_P3.

  In this embodiment, it is assumed that the first threshold value is 0.02 and the second threshold value is 0.8. In the example of FIG. 7, the first gradation value BI = 0.002 corresponds to the second gradation value HI = 0.005, and the first gradation value 0.8 corresponds to the second gradation value HI = 0.5. It is attached. Therefore, in this embodiment, the minimum gradation value ST_P2 = 0.005 and the maximum gradation value ST_P3 = 0.5 are obtained.

The first threshold value and the second threshold value are not limited to the above values. The first threshold value may be larger or smaller than 0.02. The second threshold may be larger or smaller than 0.8. For example, 0.05 may be used as the first threshold and 0.6 may be used as the second threshold. The first threshold value and the second threshold value may be fixed values determined in advance by the manufacturer, or may be values set by the user. The first threshold value and the second threshold value may be determined according to the installation environment of the image processing apparatus 100 and the display unit 107 (image display apparatus).
In addition, the determination method of a maintenance range is not restricted to the said method. For example, it corresponds to the gradation value with the smallest value among the gradation values whose cumulative frequency accumulated from the low gradation side or the high gradation side is greater than or equal to the third threshold value. The second gradation value to be determined may be determined as the minimum gradation value of the maintenance range. Then, the second gradation value corresponding to the smallest gradation value among the gradation values having the cumulative frequency equal to or greater than the fourth threshold value may be determined as the maximum gradation value of the maintenance range. The minimum gradation value and the maximum gradation value of the maintenance range may be determined according to the installation environment of the image processing apparatus 100 and the display unit 107 (image display apparatus). The minimum gradation value and the maximum gradation value of the maintenance range may be determined according to a user operation.
The third threshold value may be any value. The fourth threshold value may be any value as long as it is smaller than the third threshold value. The third threshold value and the fourth threshold value may be fixed values determined in advance by the manufacturer, or may be values set by the user. The third threshold value and the fourth threshold value may be determined according to the installation environment of the image processing apparatus 100 and the display unit 107 (image display apparatus).

The tone map conversion unit 202 generates a lookup table LUT_B based on the tone map TM_B, the minimum gradation value ST_P2, the maximum gradation value ST_P3, and the panel luminance range. The lookup table LUT_B represents the coefficient C of the first gradation value BI corresponding to the second gradation value HI for each second gradation value HI.
Then, the tone map conversion unit 202 multiplies the first gradation value by the coefficient of the first gradation value for each first gradation value represented by the tone map TM_B, thereby obtaining the panel conversion tone map P_TM_b. Generate. That is, the panel conversion tone map P_TM_b is generated by converting each first gradation value BI represented by the tone map TM_B into the display gradation value PI using the following Expression 2. In Equation 2, P_TM_b (x) is the display gradation value PI corresponding to the second gradation value x, and TM_B (x) is the first gradation value BI corresponding to the second gradation value x. Yes, LUT_B (x) is a coefficient C corresponding to the second gradation value x.

P_TM_b (x) = TM_B (x) × LUT_B (x) (Expression 2)

In this embodiment, 1 is set as the coefficient C of the first gradation value BI corresponding to the second gradation value HI belonging to the maintenance range.
Specifically, 1 is set as the coefficient C of the first gradation value BI corresponding to the second gradation value HI not less than the minimum gradation value ST_P2 and not more than the maximum gradation value ST_P3.

  Further, in this embodiment, a value larger than 1 is set as the coefficient C of the first gradation value BI corresponding to the second gradation value HI that is larger than the maximum gradation value ST_P3 of the maintenance range. Then, a value smaller than 1 is set as the coefficient C of the first gradation value BI corresponding to the second gradation value HI smaller than the minimum gradation value ST_P2 of the maintenance range.

  Specifically, as the coefficient C of the first gradation value BI corresponding to the maximum value of the second gradation value HI, the ratio of the gradation value corresponding to the maximum luminance of the panel luminance range to the first gradation value BI. Is set. As the coefficient C of the first gradation value BI corresponding to the minimum value of the second gradation value HI, the ratio of the gradation value corresponding to the minimum luminance of the panel luminance range to the first gradation value BI is set. The coefficient C of the first gradation value BI corresponding to the second gradation value HI belonging to the maintenance range, the coefficient C of the first gradation value BI corresponding to the maximum value of the second gradation value HI, and the first Interpolation processing using the coefficient C of the first gradation value BI corresponding to the minimum value of the two gradation values HI is performed. Accordingly, the first gradation that does not correspond to the second gradation value HI belonging to the maintenance range, does not correspond to the maximum value of the second gradation value HI, and does not correspond to the minimum value of the second gradation value HI. The coefficient of value BI is determined.

In the present embodiment, the maximum luminance of the panel luminance range is 200 [cd / m ² ], which is twice the luminance (100 [cd / m ² ]) corresponding to the maximum value of the first gradation value BI. . The minimum luminance of the panel luminance range is 0.1 [cd / m ² ], and the luminance (0.2 [cd / m ² ]) corresponding to the minimum value of the first gradation value BI is 0.5. Is double.
Therefore, as shown in FIG. 8, 2 is set as the coefficient C of the first gradation value BI (maximum value) corresponding to the second gradation value HI = 10, and the second gradation value HI = 0.0001. 0.5 is set as the coefficient C of the first gradation value BI (minimum value) corresponding to. Further, 1 is set as the coefficient C of the first gradation value BI corresponding to the second gradation value HI of 0.005 or more and 0.5 or less. Then, another coefficient C is determined by linear interpolation. Specifically, the other coefficient C is determined so that the coefficient C increases linearly as the second gradation value HI increases.

When the look-up table of FIG. 8 is used, the tone map TM_B of FIG. 7 is converted into the panel conversion tone map P_TM_b of FIG. When the panel conversion tone map P_TM_b in FIG. 7 is used, the gradation value of the HDR image data is converted into the gradation value of the display image data as shown below.
As the gradation value of the display image data, a larger gradation value is obtained as the gradation value of the HDR image data is larger.
The gradation value belonging to the maintenance range is converted to the same value as the first gradation value corresponding to the gradation value.
A gradation value larger than the maximum gradation value in the maintenance range is converted into a gradation value larger than the first gradation value corresponding to the gradation value.
A gradation value smaller than the minimum gradation value in the maintenance range is converted into a gradation value smaller than the first gradation value corresponding to the gradation value.
A gradation value equal to the maximum value of the second gradation value is converted into a gradation value corresponding to the maximum luminance of the panel luminance range.
A gradation value equal to the minimum value of the second gradation value is converted into a gradation value corresponding to the minimum luminance in the panel luminance range.

  Thereby, it is possible to obtain a display image whose appearance in the region where the gradation value of the HDR image data belongs to the maintenance range is equal to that of the base image. In addition, a display image having a wider dynamic range than the base image can be obtained. Specifically, the region where the gradation value of the HDR image data is larger than the maximum gradation value of the maintenance range is brighter than the base image, and the gradation value of the HDR image data is smaller than the minimum gradation value of the maintenance range. A display image in which the small area is darker than the base image can be obtained.

  As described above, according to the present embodiment, the second gradation value belonging to the maintenance range is converted to the same value as the first gradation value corresponding to the second gradation value, and the base image data is used. Display image data having a wide dynamic range. Thereby, it is possible to obtain display image data having an impression close to the image represented by the base image data and having a suitably expanded dynamic range.

  Note that the characteristics of the lookup table representing the coefficient C are not limited to the characteristics shown in FIG. For example, nonlinear interpolation may be performed to generate a look-up table having S-curve characteristics as shown in FIG.

<Example 3>
Hereinafter, an image processing apparatus and an image processing method according to Embodiment 3 of the present invention will be described.
In the first and second embodiments, the reverse tone map is acquired from the outside, and the panel conversion tone map is generated from the reverse tone map.
In the present embodiment, an example will be described in which reverse ratio data is acquired from the outside, and panel conversion ratio data is generated from the reverse ratio data. The inverse ratio data is ratio data that represents the ratio between the gradation value of the HDR image data and the gradation value of the base image data for each image region composed of one or a plurality of pixels. In this embodiment, an example in which the ratio represented by the inverse ratio data is the ratio of the gradation value of the HDR image data to the gradation value of the base image data will be described. However, the ratio represented by the inverse ratio data is the HDR image. It may be the ratio of the gradation value of the base image data to the gradation value of the data.

FIG. 10 is a block diagram illustrating an example of a functional configuration of the image processing apparatus 300 according to the present embodiment.
The image processing apparatus 300 includes a decode processing unit 302, a ratio data generation unit 303, a maintenance range determination unit 304, a ratio data conversion unit 305, a range compression processing unit 306, a display unit 107, and the like.
In FIG. 10, the same reference numerals as those in FIGS. 1 and 6 are assigned to the same functional units as those in the first and second embodiments (FIGS. 1 and 6), and the description thereof is omitted.
In this embodiment, it is assumed that the panel luminance range, the dynamic range of base image data, the dynamic range of HDR image data, and the dynamic range of display image data are the same as those in the first embodiment.

An HDR file 301 is input to the image processing apparatus 300.
The HDR file 301 is a data file including base image data (first image data) and corresponding data (gradation difference data).
In this embodiment, reverse ratio data Ratio is used as the corresponding data.

  The decode processing unit 302 generates (restores; acquires) HDR image data using the base image data and the reverse ratio data Ratio included in the HDR file 301 (first acquisition process). Then, the decode processing unit 102 outputs the generated HDR image data.

The ratio data generation unit 303 generates (acquires) ratio data Ratio_C based on the reverse ratio data Ratio included in the HDR file 301 (second acquisition process). Then, the ratio data generation unit 303 outputs the generated ratio data Ratio_C.
The ratio data Ratio_C is first correspondence data representing the correspondence between the first gradation value and the second gradation value. Specifically, the ratio data Ratio_C is first ratio data that represents, for each image region, a first ratio that is a ratio between the gradation value of the HDR image data and the gradation value of the base image data. The first ratio represented by the ratio data Ratio_C is the reciprocal of the ratio represented by the inverse ratio data Ratio. In the present embodiment, the first ratio is the ratio of the gradation value of the base image data to the gradation value of the HDR image data.

The maintenance range determination unit 304 determines a partial range of the second gradation value range as the maintenance range. Then, the maintenance range determination unit 304 outputs maintenance range information representing the determined maintenance range. In this embodiment, two second gradation values corresponding to the extreme value of the first ratio in the change in the first ratio with respect to the change in the second gradation value are the maximum gradation value and the minimum gradation value of the maintenance range. It is determined.
In addition, the determination method of a maintenance range is not restricted to the said method. For example, the maintenance range may be determined by the same method as in the first and second embodiments.

The ratio data conversion unit 305 generates panel conversion ratio data P_Ratio_c based on the ratio data Ratio_C, the maintenance range, and the panel luminance range. In the present embodiment, the panel conversion ratio data P_Ratio_c is generated by correcting each first ratio represented by the ratio data Ratio_C. Then, the ratio data conversion unit 305 outputs the generated panel conversion ratio data P_Ratio_c.
The panel conversion ratio data P_Ratio_c is second correspondence data representing the correspondence between the second gradation value and the display gradation value. Specifically, the panel conversion ratio data P_Ratio_c is second ratio data that represents a second ratio, which is a ratio between the gradation value of the HDR image data and the gradation value of the display image data, for each image area. is there. In the present embodiment, the second ratio is a ratio of the gradation value of the display image data to the gradation value of the HDR image data.

The range compression processing unit 306 uses the panel conversion ratio data P_Ratio_c generated by the ratio data conversion unit 305 to convert the HDR image data generated by the decoding processing unit 302 into display image data. Then, the range compression processing unit 306 outputs display image data. In this embodiment, display image data is generated by compressing the dynamic range of the HDR image data so that the luminance range corresponding to the range of gradation values that can be taken by the image data matches the panel luminance range. .
Note that the luminance range corresponding to the range of gradation values that the display image data can take may be narrower than the panel luminance range.

FIG. 11 shows an example of the ratio data Ratio_C and the panel conversion ratio data P_Ratio_c. FIG. 11 also shows a tone map that is not generated in this embodiment and a tone map for panel conversion. The thick solid line in FIG. 11 indicates ratio data Ratio_C, and the thick broken line in FIG. 11 indicates panel conversion ratio data P_Ratio_c. A thin solid line in FIG. 11 indicates a tone map, and a thin broken line in FIG. 11 indicates a panel conversion tone map. The horizontal axis of FIG. 11 indicates the gradation value of HDR image data (tone map and panel conversion tone map input value (second gradation value)). The vertical axis on the left side of FIG. 11 indicates the output values (first gradation value and display gradation value) of the tone map and panel conversion tone map, and the vertical axis on the right side of FIG. 11 indicates the ratio (first ratio). And the second ratio).
FIG. 11 shows ratio data Ratio_C and a tone map for reducing the gradation of the second gradation value range (bright portion) on the high gradation side and the second gradation value range (dark portion) on the low gradation side. It is shown.

  The maintenance range determination unit 304, the ratio data conversion unit 305, and the range compression processing unit 306 will be described in more detail with reference to FIGS.

The maintenance range determination unit 304 uses the ratio data Ratio_C to determine the maximum gradation value and the minimum gradation value of the maintenance range.
Specifically, the maintenance range determination unit 304 uses the ratio data Ratio_C to grasp the correspondence between the second gradation value and the first ratio. That is, the maintenance range determination unit 304 grasps the correspondence relationship indicated by the thick solid line in FIG. The process for grasping the correspondence is, for example, a process for calculating a function representing the correspondence. Then, based on the above correspondence relationship, the maintenance range determination unit 304 determines the two second gradation values corresponding to the extreme values of the first ratio in the change in the first ratio with respect to the change in the second gradation value as the maintenance range. The maximum gradation value and the minimum gradation value are determined. Specifically, the second gradation value that is the minimum value and smaller than 1 is determined as the minimum gradation value ST_P4 of the maintenance range. Then, the second gradation value that is the maximum value and larger than 1 is determined as the maximum gradation value ST_P5 of the maintenance range. In the example of FIG. 11, 0.01 is obtained as the minimum gradation value ST_P4, and 0.5 is obtained as the maximum gradation value ST_P5 of the maintenance range.
Then, the maintenance range determination unit 304 outputs the minimum gradation value ST_P4 and the maximum gradation value ST_P5.

The ratio data conversion unit 305 generates a lookup table LUT_C based on the ratio data Ratio_C, the minimum gradation value ST_P4, the maximum gradation value ST_P5, and the panel luminance range. The lookup table LUT_C represents a first ratio coefficient C corresponding to the second gradation value HI for each second gradation value HI.
Then, the ratio data conversion unit 305 generates panel conversion ratio data P_Ratio_c by multiplying the first ratio by the coefficient of the first ratio for each first ratio represented by the ratio data Ratio_C. That is, the panel conversion ratio data P_Ratio_c is generated by converting each first ratio represented by the ratio data Ratio_C into the second ratio using the following Expression 3. In Expression 3, P_Ratio_c (x) is a second ratio corresponding to the second gradation value x, and Ratio_C (x) is a first ratio corresponding to the second gradation value x, and LUT_C (x) Is a coefficient C corresponding to the second gradation value x.

P_Ratio_c (x) = Ratio_C (x) × LUT_C (x)
... (Formula 3)

In this embodiment, 1 is set as the first ratio coefficient C corresponding to the second gradation value HI belonging to the maintenance range.
Specifically, 1 is set as the coefficient C of the first ratio corresponding to the second gradation value HI not less than the minimum gradation value ST_P4 and not more than the maximum gradation value ST_P5.

  In this embodiment, a value larger than 1 is set as the first ratio coefficient C corresponding to the second gradation value HI that is larger than the maximum gradation value ST_P5 of the maintenance range. Then, a value smaller than 1 is set as the first ratio coefficient C corresponding to the second gradation value HI smaller than the minimum gradation value ST_P4 of the maintenance range.

  Specifically, as the first ratio coefficient C corresponding to the maximum value of the second gradation value HI, the first gradation corresponding to the first ratio of the gradation value corresponding to the maximum brightness of the panel brightness range. A ratio to the value BI is set. As the first ratio coefficient C corresponding to the minimum value of the second gradation value HI, the ratio of the gradation value corresponding to the minimum brightness of the panel brightness range to the first gradation value BI corresponding to the first ratio is Is set. The first ratio coefficient C corresponding to the second gradation value HI belonging to the maintenance range, the first ratio coefficient C corresponding to the maximum value of the second gradation value HI, and the second gradation value HI Interpolation processing using the first ratio coefficient C corresponding to the minimum value is performed. Accordingly, the first ratio does not correspond to the second gradation value HI belonging to the maintenance range, does not correspond to the maximum value of the second gradation value HI, and does not correspond to the minimum value of the second gradation value HI. A coefficient is determined.

In the present embodiment, the maximum luminance of the panel luminance range is 200 [cd / m ² ], which is twice the luminance (100 [cd / m ² ]) corresponding to the maximum value of the first gradation value BI. . The minimum luminance of the panel luminance range is 0.1 [cd / m ² ], and the luminance (0.2 [cd / m ² ]) corresponding to the minimum value of the first gradation value BI is 0.5. Is double.
Therefore, as shown in FIG. 12, 2 is set as the first ratio coefficient C corresponding to the second gradation value HI = 10, and the first ratio corresponding to the second gradation value HI = 0.0001 is set. As the coefficient C, 0.5 is set. Further, 1 is set as the coefficient C of the first ratio corresponding to the second gradation value HI of 0.01 or more and 0.5 or less. Then, another coefficient C is determined by linear interpolation. Specifically, the other coefficient C is determined so that the coefficient C increases linearly as the second gradation value HI increases.

  The range compression processing unit 306 converts the gradation value of the HDR image data into the gradation value of the display image data using the panel conversion ratio data P_Ratio_c. Thereby, the HDR image data is converted into display image data. Specifically, for each pixel, the gradation value of the HDR image data of the pixel is multiplied by the second ratio associated with the image area including the pixel, thereby obtaining the gradation of the display image data of the pixel. A value is calculated. When the second ratio cr is associated with an image region including a pixel whose HDR image data gradation value is xr, xr × cr is calculated as the gradation value of the display image data of the pixel. The

When the look-up table of FIG. 12 is used, the ratio data Ratio_C of FIG. 11 is converted into the panel conversion ratio data P_Ratio_c of FIG. When the panel conversion ratio data P_Ratio_c in FIG. 11 is used, the gradation value of the HDR image data is converted into the gradation value of the display image data as shown below.
As the gradation value of the display image data, a larger gradation value is obtained as the gradation value of the HDR image data is larger.
The gradation value belonging to the maintenance range is converted to the same value as the first gradation value corresponding to the gradation value.
A gradation value larger than the maximum gradation value in the maintenance range is converted into a gradation value larger than the first gradation value corresponding to the gradation value.
A gradation value smaller than the minimum gradation value in the maintenance range is converted into a gradation value smaller than the first gradation value corresponding to the gradation value.
A gradation value equal to the maximum value of the second gradation value is converted into a gradation value corresponding to the maximum luminance of the panel luminance range.
A gradation value equal to the minimum value of the second gradation value is converted into a gradation value corresponding to the minimum luminance in the panel luminance range.

  Thereby, it is possible to obtain a display image whose appearance in the region where the gradation value of the HDR image data belongs to the maintenance range is equal to that of the base image. In addition, a display image having a wider dynamic range than the base image can be obtained. Specifically, the region where the gradation value of the HDR image data is larger than the maximum gradation value of the maintenance range is brighter than the base image, and the gradation value of the HDR image data is smaller than the minimum gradation value of the maintenance range. A display image in which the small area is darker than the base image can be obtained.

A specific example of processing of the image processing apparatus according to the present embodiment will be described with reference to FIG.
FIG. 13 shows ratio data Ratio_C, panel conversion ratio data P_Ratio_c, and each image. FIG. 13 also shows a tone map that is not generated in this embodiment and a panel conversion tone map tone map. In the upper part on the right side of FIG. 13, an HDR image a which is an example of an HDR image is shown. In the middle part on the right side of FIG. 13, a base image b, which is an example of a base image, is shown. A display image c, which is an example of a display image, is shown in the lower part on the right side of FIG. The numerical value described in each image is a gradation value in an area where the numerical value is described.

  In the present embodiment, the base image b is converted into the HDR image a using the inverse ratio data that represents the inverse of the ratio represented by the ratio data Ratio_C as the ratio. Then, the ratio data Ratio_C is converted into panel conversion ratio data P_Ratio_c. Thereafter, the HDR image a is converted into the display image c using the panel conversion ratio data P_Ratio_c.

In the ratio data Ratio_C in FIG. 13, the first ratio 1 is the second gradation value 0.1, the first ratio 0.45 is the second gradation value 2, and the first ratio is the second gradation value 0.001. 3 is associated. Therefore, when the gradation value of the HDR image a is converted using the ratio data Ratio_C in FIG. 13, the background gradation value 0.1 is maintained at the gradation value 0.1 (0.1 × 1). ). The left square tone value 2 is converted to a tone value of 0.9 (= 2 × 0.45), and the right square tone value 0.001 is converted to a tone value of 0.003 (= 0.001 × 3). ). As described above, when the gradation value of the HDR image a is converted using the ratio data Ratio_C, a gradation value equal to the gradation value of the base image b is obtained as the converted gradation value.

  In the panel conversion ratio data P_Ratio_c in FIG. 13, the second gradation value 0.1 is associated with the second ratio 0.1 that is the same as the first ratio. Further, the second gradation value 2 is associated with a second ratio 0.9 greater than the first ratio 1.0, and the second gradation value 0.001 is 1 smaller than the first ratio 3. .5 are associated with each other. Therefore, when the gradation value of the HDR image a is converted using the panel conversion ratio data P_Ratio_c in FIG. 13, the background gradation value 0.1 is converted to the same gradation value as that of the base image b. Specifically, the gradation value 0.1 of the background is maintained as the gradation value 0.1 (0.1 × 1). The left square tone value 2 is converted to a brighter tone value 1.8 (= 2 × 0.9) than the base image b, and the right square tone value 0.001 is darker than the base image b. The tone value is converted to 0.0015 (= 0.001 × 1.5). Thereby, the HDR image a is converted into the display image c. As a result, it is possible to obtain a display image in which the appearance of the background is equal to the base image, the left square is brighter than the base image, and the right square is darker than the base image.

  As described above, according to the present embodiment, HDR image data is converted into display image data using the ratio data as in the first and second embodiments. Thereby, it is possible to obtain display image data having an impression close to the image represented by the base image data and having a suitably expanded dynamic range.

  As mentioned above, although preferable Example 1-3 of this invention was demonstrated, this invention is not limited to Examples 1-3, A various deformation | transformation and change are possible within the range of the summary. For example, the present invention may be modified or changed as appropriate according to the configuration of the apparatus to which the present invention is applied and various conditions. Moreover, it is also possible to combine suitably the structure of Examples 1-3.

<Other examples>
The present invention can also be implemented by a system (or a device such as a CPU or MPU) of a system or apparatus that implements the functions of the above-described embodiments by reading and executing a program recorded in a storage device. The present invention can also be implemented by a method comprising steps executed by a computer of a system or apparatus that implements the functions of the above-described embodiments by reading and executing a program recorded in a storage device, for example. . For this purpose, the program is stored in the computer from, for example, various types of recording media that can serve as the storage device (ie, computer-readable recording media that holds data non-temporarily). Provided to. Therefore, the computer (including devices such as CPU and MPU), the method, the program (including program code and program product), and the computer-readable recording medium that holds the program non-temporarily are all present. It is included in the category of the invention.

100, 200, 300: Image processing apparatus 102, 302: Decoding processing unit 103: Tone map generation unit 104, 201, 304: Maintenance range determination unit 105, 202: Tone map conversion unit 106, 306: Range compression processing unit 303: Ratio data generation unit 305: Ratio data conversion unit

Claims (20)

First acquisition means for acquiring second image data with an expanded dynamic range of the first image data;
First correspondence data representing a correspondence relationship between a first gradation value that is a gradation value that can be taken by the first image data and a second gradation value that is a gradation value that can be taken by the second image data. A second acquisition means for acquiring;
Determining means for determining a partial range of the second gradation value range as a maintenance range;
Based on the first correspondence data, the maintenance range, and a luminance range that can be taken by the image displayed on the screen of the display unit, the second gradation value and the display image data are taken. Generating means for generating second correspondence data representing a correspondence relationship between display gradation values that are obtained gradation values;
Conversion means for converting the second image data into the display image data using the second correspondence data;
Have
The luminance range is wider than the luminance range corresponding to the first gradation value range and narrower than the luminance range corresponding to the second gradation value range;
The generation unit converts the second gradation value belonging to the maintenance range into the same value as the first gradation value corresponding to the second gradation value, and has a dynamic range larger than that of the first image data. The image processing apparatus, wherein the second correspondence data is generated so that the wide display image data is generated. The generating means includes
A gradation value corresponding to the maximum luminance of the luminance range, wherein the second gradation value larger than the maximum gradation value of the maintenance range is larger than the first gradation value corresponding to the second gradation value. Is converted to the following value,
A gradation value that is smaller than the first gradation value corresponding to the second gradation value and whose second gradation value is smaller than the minimum gradation value of the maintenance range and that corresponds to the minimum luminance of the luminance range. As converted to the above values,
The image processing apparatus according to claim 1, wherein the second correspondence data is generated. The generating means includes
The maximum value of the second gradation value is converted into a gradation value corresponding to the maximum luminance of the luminance range;
The minimum value of the second gradation value is converted into a gradation value corresponding to the minimum luminance of the luminance range,
The image processing apparatus according to claim 1, wherein the second correspondence data is generated. The first correspondence data is a first lookup table that represents the second gradation value as an input value, and represents the first gradation value as an output value;
4. The second correspondence data is a second lookup table that represents the second gradation value as an input value and represents the display gradation value as an output value. The image processing apparatus according to any one of the above. The image processing apparatus according to claim 4, wherein the generation unit generates the second lookup table by correcting each first gradation value represented by the first lookup table. 6. The image according to claim 4, wherein the generation unit generates the second lookup table by multiplying each first gradation value represented by the first lookup table by a coefficient. Processing equipment. The generating means includes
Using a ratio of the gradation value corresponding to the maximum luminance of the luminance range to the first gradation value as a coefficient of the first gradation value corresponding to the maximum value of the second gradation value;
The ratio of the gradation value corresponding to the minimum luminance of the luminance range to the first gradation value is used as the coefficient of the first gradation value corresponding to the minimum value of the second gradation value. The image processing apparatus according to claim 6. The first correspondence data represents a first ratio, which is a ratio between the gradation value of the second image data and the gradation value of the first image data, for each image region composed of one or a plurality of pixels. 1 ratio data,
The second correspondence data is second ratio data representing a second ratio, which is a ratio between the gradation value of the second image data and the gradation value of the display image data, for each image area. The image processing apparatus according to claim 1, wherein the image processing apparatus is characterized in that: The image processing apparatus according to claim 8, wherein the generation unit generates the second ratio data by correcting each first ratio represented by the first ratio data. The image processing apparatus according to claim 8, wherein the generation unit generates the second ratio data by multiplying each first ratio represented by the first ratio data by a coefficient. The generating means includes
As a coefficient of the first ratio corresponding to the maximum value of the second gradation value, a ratio of the gradation value corresponding to the maximum luminance of the luminance range to the first gradation value corresponding to the first ratio is used. ,
The ratio of the gradation value corresponding to the minimum luminance of the luminance range to the first gradation value corresponding to the first ratio is used as the coefficient of the first ratio corresponding to the minimum value of the second gradation value. The image processing apparatus according to claim 10. The determining means determines one second gradation value as a maintenance gradation value,
The generating means includes
A second gradation value equal to the sustain gradation value is converted to the same value as the first gradation value corresponding to the second gradation value;
A second gradation value larger than the sustain gradation value is larger than the first gradation value corresponding to the second gradation value and is equal to or smaller than the gradation value corresponding to the maximum luminance in the luminance range. Is converted to
The second gradation value smaller than the sustain gradation value is smaller than the first gradation value corresponding to the second gradation value and is equal to or larger than the gradation value corresponding to the minimum luminance in the luminance range. To be converted to
The image processing apparatus according to claim 1, wherein the second correspondence data is generated. The image processing apparatus according to claim 12, wherein the determining unit determines a second gradation value corresponding to an intermediate value of the first gradation values as the sustain gradation value. The image processing apparatus according to claim 12, wherein the determination unit determines a second gradation value corresponding to an intermediate value of gradation values of the first image data as the sustain gradation value. The image processing apparatus according to claim 1, wherein the determination unit determines a maximum gradation value and a minimum gradation value of the maintenance range. The determining means includes
Determining a second gradation value corresponding to a first gradation value equal to the first threshold as a minimum gradation value of the maintenance range;
16. The image according to claim 15, wherein a second gradation value corresponding to a first gradation value equal to a second threshold value greater than the first threshold value is determined as a maximum gradation value of the maintenance range. Processing equipment. The determining means includes
Corresponds to the gradation value with the smallest value among the gradation values whose cumulative frequency accumulated from the low gradation side or the high gradation side is greater than or equal to the third threshold value, expressed by the histogram of gradation values of the first image data. Determining the second gradation value to be the minimum gradation value of the maintenance range;
The second gradation value corresponding to the smallest gradation value among gradation values whose cumulative frequency is greater than or equal to the fourth threshold value greater than the third threshold value is determined as the maximum gradation value of the maintenance range. The image processing apparatus according to claim 15, wherein: The determining means determines two second gradation values corresponding to extreme values of the ratio in a change in a ratio of the second gradation value and the first gradation value with respect to a change in the second gradation value. 16. The image processing apparatus according to claim 15, wherein the maximum gradation value and the minimum gradation value of the maintenance range are determined. A first acquisition step of acquiring second image data with an expanded dynamic range of the first image data;
First correspondence data representing a correspondence relationship between a first gradation value that is a gradation value that can be taken by the first image data and a second gradation value that is a gradation value that can be taken by the second image data. A second obtaining step to obtain;
A determination step of determining a partial range of the second gradation value range as a maintenance range;
Based on the first correspondence data, the maintenance range, and a luminance range that can be taken by the image displayed on the screen of the display unit, the second gradation value and the display image data are taken. A generation step of generating second correspondence data representing a correspondence relationship with the display gradation value which is a gradation value to be obtained;
A conversion step of converting the second image data into the display image data using the second correspondence data;
Have
The luminance range is wider than the luminance range corresponding to the first gradation value range and narrower than the luminance range corresponding to the second gradation value range;
In the generating step, the second gradation value belonging to the maintenance range is converted to the same value as the first gradation value corresponding to the second gradation value, and the dynamic range is greater than that of the first image data. The image processing method, wherein the second correspondence data is generated so that the wide display image data is generated.   A program causing a computer to execute each step of the image processing method according to claim 19.

Images