JP2019110538A - Luminance conversion method, luminance conversion device, and video display device - Google Patents

Abstract

A luminance conversion method, a luminance conversion device, and an image display device for improving a displayable luminance level are provided. A luminance conversion method for converting the luminance of a video, wherein a first luminance signal, which is a luminance signal of the video, is acquired, and a first code value included in the first luminance signal is used as a luminance in a first luminance range. The first luminance value is converted into the first luminance value using a first EOTF (Electro-Optical Transfer Function) indicating the correspondence between the value and the code value, and the first luminance value is preliminarily related to the first luminance value. And the second luminance value is converted into a second code value using a second EOTF indicating a correspondence relationship between the luminance value and the code value of the second luminance range having a maximum value different from the first luminance range, and the second A second luminance signal including the code value is output. [Selection] Figure 9

Description

  The present disclosure relates to a luminance conversion method, a luminance conversion device, and a video display device that convert signals into different luminance ranges.

  Conventionally, an image signal processing apparatus for improving the displayable luminance level is disclosed (see, for example, Patent Document 1).

JP, 2008-167418, A

  A luminance conversion method according to an aspect of the present disclosure is a luminance conversion method for converting the luminance of a video, wherein a first luminance signal that is a luminance signal of the video is acquired, and a first luminance signal included in the first luminance signal. The code value is converted into a first brightness value using a first EOTF (Electro-Optical Transfer Function) indicating a correspondence between the brightness value of the first brightness range and the code value, and the first brightness value is converted to the first brightness value. A second luminance value previously associated with the luminance value, and converting the second luminance value to a correspondence between the luminance value of the second luminance range having a different maximum value from the first luminance range and the code value; 2. Convert to a second code value using 2EOTF, and output a second luminance signal including the second code value.

  In addition, a conversion method according to another aspect of the present disclosure is a conversion method regarding luminance of an image, wherein the luminance of the image includes luminance values in a first luminance range, and the luminance value of the image is quantized. To obtain a first luminance signal indicating a code value obtained by performing a quantization on a second luminance range having a maximum value different from the first luminance range from the code value indicated by the acquired first luminance signal The associated code value is determined as a converted code value, and the first luminance signal is converted into a second luminance signal indicating the converted code value.

  According to the above aspect, further improvement can be realized.

FIG. 1 is a diagram showing an example of an EOTF (Electro-Optical Transfer Function). FIG. 2 is an explanatory diagram of a method of determining the code value of the luminance signal stored in the content and a process of restoring the luminance value from the code value at the time of reproduction. FIG. 3 is an explanatory view of a production of a BD and a player for reproducing the BD. FIG. 4A is an example in which a BD player and a TV are connected by HDMI (registered trademark), and a case in which the TV supports HDR display. FIG. 4B is an example in which a BD player and a TV are connected by HDMI (registered trademark), and a case in which the TV does not correspond to HDR display is illustrated. FIG. 5A is a diagram for describing an example in which the SDR signal is first remapped to the HDR signal. FIG. 5B is a diagram for describing an example of first remapping the HDR signal to the SDR signal. FIG. 6 shows the luminance range of the SDR master and the HDR master in video. FIG. 7 is a diagram for describing an example of relationship information that is a correspondence between the luminance value of HDR and the luminance value of SDR when converting an HDR signal into an SDR signal. FIG. 8 is a block diagram showing a configuration of a remapping unit in the conversion device. FIG. 9 is a diagram showing a flowchart of remapping processing in the conversion device. FIG. 10 is a diagram showing an example of combination of HDR and SDR in the case where one content of each of video and graphics streams is included in content. FIG. 11 illustrates generating a graphics master using a common EOTF with a video master. FIG. 12A is a diagram for describing a case of mapping to an SDR signal in generation of a graphics master. FIG. 12B is a diagram for describing a case of mapping to an HDR signal in generation of a graphics master. FIG. 13 is a block diagram showing a configuration of a generation unit that generates a graphics signal in authoring. FIG. 14 is a flowchart showing a method of generating a graphics signal in authoring. FIG. 15 shows the BT. 709 and BT. It is the figure which showed the color space of 709 with a CIE colorimetric system.

(Findings that formed the basis of this disclosure)
The present inventor has found that the following problems occur in the image signal processing apparatus described in the "Background Art" section.

  In the image signal processing apparatus disclosed in Patent Document 1, linear luminance is calculated for each pixel based on linear RGB values calculated from pixels constituting an object, and for each pixel based on linear RGB values and linear luminance. The correction linear luminance and the correction linear RGB value of the composite pixel combining the plurality of pixels including the pixel are calculated, and the correction linear luminance and the correction linear RGB value are gamma corrected to calculate the display luminance and the display RGB value. . As described above, in the image signal processing apparatus, the number of displayable gradations is increased by correcting the linear luminance based on the corrected linear RGB values.

  However, in the correction (conversion) of the luminance of the image signal processing apparatus disclosed in Patent Document 1, the luminance is corrected (converted) to the second luminance range in which the luminance range is expanded or reduced from the first luminance range. The conversion method of the luminance was not considered.

  Based on the above examination, the inventor examined the following improvement measures in order to solve the above-mentioned problems.

  A luminance conversion method according to an aspect of the present disclosure is a luminance conversion method for converting the luminance of a video, wherein a first luminance signal that is a luminance signal of the video is acquired, and a first luminance signal included in the first luminance signal. The code value is converted into a first brightness value using a first EOTF (Electro-Optical Transfer Function) indicating a correspondence between the brightness value of the first brightness range and the code value, and the first brightness value is converted to the first brightness value. A second luminance value previously associated with the luminance value, and converting the second luminance value to a correspondence between the luminance value of the second luminance range having a different maximum value from the first luminance range and the code value; 2. Convert to a second code value using 2EOTF, and output a second luminance signal including the second code value.

  Further, when the first luminance value is in a low luminance area of the first luminance range, the first luminance value is converted into the second luminance value having a luminance value substantially equal to the first luminance value, When the first brightness value is in a high brightness area of the first brightness range, the second brightness is such that the increase amount of the brightness value decreases as the first brightness value increases. It may be converted to a value.

  In addition, when the first luminance value exceeds the maximum luminance value of the second luminance range, the first luminance value is converted to the second luminance value which is the maximum luminance value of the second luminance range. It is also good.

  Further, the method of converting the first luminance value to the second luminance value may be determined according to the scene of the video.

  In addition, a conversion method according to another aspect of the present disclosure is a conversion method regarding luminance of an image, wherein the luminance of the image includes luminance values in a first luminance range, and the luminance value of the image is quantized. To obtain a first luminance signal indicating a code value obtained by performing a quantization on a second luminance range having a maximum value different from the first luminance range from the code value indicated by the acquired first luminance signal The associated code value is determined as a converted code value, and the first luminance signal is converted into a second luminance signal indicating the converted code value.

  According to this, it is possible to appropriately convert the luminance from the first luminance range to the second luminance range in which the luminance range is expanded or reduced.

  Also, for example, the first luminance signal may be a luminance value of the image using a first EOTF (Electro-Optical Transfer Function) in which the luminance value in the first luminance range is associated with a plurality of first code values. Represents the first code value obtained by the quantization, and in the conversion to the second luminance signal, the first EOTF, the luminance value in the second luminance range, and a plurality of second code values From the code value indicated by the acquired first luminance signal using the associated second EOTF, the corresponding second code value is determined as a converted code value, and the second luminance signal is determined The second code value may be indicated.

  Also, for example, in the conversion to the second luminance signal, (i) using the first EOTF, determine a luminance value related to a code value indicated by the first luminance signal, and (ii) A first remapping may be performed on the determined luminance value to determine the second code value associated in the second EOTF as a converted code value.

  Also, for example, in the first remap, in the second EOTF, when there is no code value in which the determined luminance value is associated with the plurality of second code values, among the plurality of second code values, Then, the code value associated with the luminance value having the smallest difference from the determined luminance value may be determined as the converted code value.

  Also, for example, in the first remap, the second bit number is smaller than the first bit number at which the first code value represented by the first EOTF is associated with the code value indicated by the acquired first luminance signal. In some cases, the luminance value associated with the first EOTF may be determined by using the upper bits by the second bit number of the first code value.

  Also, for example, in the conversion to the second luminance signal, (i) using the first EOTF, determine a first luminance value related to a code value indicated by the first luminance signal, (ii) Determining a second luminance value in the second luminance range, which is previously related to the determined first luminance value, and (iii) regarding the determined second luminance value, in the second EOTF A second remapping may be performed to determine the second code value as a post-conversion code value.

  Further, for example, the first luminance range is a luminance range in which the maximum luminance value is larger than the second luminance range, and the maximum luminance value in the first luminance range and the maximum luminance value in the second luminance range are In the determination of the second luminance value, when the determined first luminance value is in the low luminance region on the side where the luminance is low in the first luminance range, the second luminance value is substantially the same as the first luminance value. The second luminance value is determined to be equal luminance values, and the first luminance value is determined if the determined first luminance value is in the high luminance region on the high luminance side of the first luminance range. The second brightness value is determined such that the increase amount of the brightness value decreases as the value of the second brightness value increases, and the second brightness value is determined when the determined first brightness value is the maximum brightness value of the first brightness range. The maximum luminance value of the range may be the second luminance value.

  Also, for example, in the determination of the second luminance value, when the determined first luminance value exceeds the maximum luminance value of the second luminance range, the maximum luminance value of the second luminance range is the second luminance value. It may be a luminance value.

  Also, for example, in the determination of the second luminance value, the relation according to the scene of the video from a plurality of relation information indicating the relation between the luminance value in the first luminance range and the luminance value in the second luminance range Information may be selected, and the second luminance value may be determined from the determined first luminance value using the selected relationship information.

  Also, for example, in the conversion to the second luminance signal, when the video is a video and the first luminance signal is a signal obtained by quantizing the luminance value of the video, (i) using the first EOTF To determine a first luminance value associated with the code value indicated by the first luminance signal, and (ii) predetermining the determined first luminance value in the second luminance range. A second remapping is performed, wherein a luminance value is determined and (iii) a code value related in the second EOTF is determined as a converted code value for the determined second luminance value, and the image is graphics If the first luminance signal is a signal obtained by quantizing the luminance value of the graphics, (i) using the first EOTF, the luminance associated with the code value indicated by the first luminance signal Determining the value, and (ii) determining the code value associated in the second EOTF as the converted code value for the determined luminance value, performing the first remapping to obtain the second luminance signal It may be converted to

  Also, for example, the video and graphics converted into the second luminance signal may be synthesized and output by performing the first remapping and the second remapping.

  Also, for example, the conversion to the second luminance signal uses the first EOTF and the second EOTF whose luminance range can be displayed on the display device to which the second luminance signal is output. May be done.

  Furthermore, for example, the second luminance signal converted from the acquired first luminance signal may be output together with the meta information for identifying the second EOTF.

  Note that these general comprehensive or specific aspects may be realized by an apparatus, a system, an integrated circuit, a computer program, or a recording medium such as a computer program or a computer readable CD-ROM, a system, a method, and an integrated circuit. , And may be realized by any combination of computer programs or recording media.

  Hereinafter, the conversion method and conversion apparatus according to an aspect of the present disclosure will be specifically described with reference to the attached drawings.

  The embodiments described below each show one specific example of the present disclosure. Numerical values, shapes, materials, components shown in the following embodiments. The arrangement positions and connection forms of the components, the steps, the order of the steps, and the like are merely examples, and are not intended to limit the present disclosure. Further, among the components in the following embodiments, components not described in the independent claim indicating the highest concept are described as arbitrary components.

Embodiment 1
[1-1. background]
Up until now, the main focus has been on expanding the number of pixels as video quality improvement, and 1920 x 1080 pixel video called Full HD (FHD: Full High Definition) or 2048 x 1080 pixel video becomes widespread It has In recent years, introduction of so-called 4K video, such as 3840 × 1920 pixels or 4096 × 1920 pixels, has been started in order to further improve the image quality of the video. Further, it has been considered to improve the image quality of the image by increasing the resolution of the image, expanding the dynamic range and the color gamut, or improving the frame rate.

  Among them, with regard to the dynamic range, the maximum luminance is used to express bright light such as specular reflection light that can not be expressed by the current TV signal while maintaining dark gradation in the conventional video, with brightness closer to reality. HDR (High Dynamic Range) has attracted attention as a method corresponding to the expanded luminance range. Specifically, the method of luminance range to which conventional TV signals are compatible is called SDR (Standard Dynamic Range), and the maximum luminance value is 100 nit, while HDR has a maximum of 1000 nit or more. It is assumed that the luminance value is enlarged. HDR is being standardized in, for example, the Society of Motion Picture & Television Engineers (SMPTE) and the International Telecommunications Union Radiocommunications Sector (ITU-R). Broadcast and BD (Blu-ray (registered trademark) Disc) etc. are assumed as a concrete application place of HDR.

[1-2. About EOTF]
Here, the EOTF will be described with reference to FIG.

  FIG. 1 is a diagram showing an example of an EOTF (Electro-Optical Transfer Function) corresponding to each of HDR and SDR.

  EOTF is generally called a gamma curve, which indicates the correspondence between luminance values and code values, and quantizes the luminance values to convert them into code values. That is, the EOTF is relationship information indicating the correspondence between the luminance value and a plurality of code values. For example, when the luminance value of an image corresponding to SDR is expressed by a code value of 8-bit gradation, the luminance value in the luminance range up to 100 nit is quantized and mapped to 256 integer values of 0 to 255 Be done. That is, by quantizing based on EOTF, luminance values (luminance values of video corresponding to SDR) in a luminance range up to 100 nit are converted into SDR signals which are code values of 8 bits. In the EOTF compatible with HDR (hereinafter referred to as "HDR EOTF"), it is possible to express a luminance value higher than that of SDR compatible EOTF (hereinafter referred to as "SDR EOTF"), For example, in FIG. 1, the maximum value of luminance (peak luminance) is 1000 nits. That is, the luminance range of HDR includes all the luminance range of SDR, and the peak luminance of HDR is larger than the peak luminance of SDR. The luminance range of the HDR is the luminance range obtained by expanding the maximum value from, for example, 100 nit, which is the maximum value of the SDR luminance range, to 1000 nit. Further, the HDR signal is expressed by, for example, 10-bit gradation.

[1-3. How to use EOTF]
FIG. 2 is an explanatory diagram of a method of determining the code value of the luminance signal stored in the content, and a process of restoring the luminance value from the code value at the time of reproduction.

  The luminance signal indicating the luminance in the present example is an HDR signal corresponding to HDR. The image after grading is quantized by the inverse function of the EOTF of HDR, and the code value corresponding to the luminance value of the image is determined. Image coding or the like is performed based on this code value to generate elementary streams of video and graphics, respectively. At the time of reproduction, the luminance value for each pixel is restored by inversely quantizing the decoding result of the elementary stream based on EOTF of HDR.

[1-4. BD stream configuration]
It was mentioned earlier that HDR may be used in optical disks such as BD or in broadcasting. Hereinafter, a BD as an example of a medium in which HDR is used will be described with reference to FIG.

  FIG. 3 is an explanatory view of a production of a BD and a player for reproducing the BD.

  As shown in FIG. 3, the production process includes authoring of Blu-ray (registered trademark) content, creation of a BD storing the authored Blu-ray (registered trademark) content, and the like. For Blu-ray (registered trademark) content, in addition to video and audio, graphics data for generating subtitles and menus, and scenario data for providing menus and interactivity in user operations, etc. Is included. Scenario data includes a format called HDMV (High Definition Movie) controlled by a prescribed command, and BD-J (Blu-ray (registered trademark) Disc Java (registered trademark)) controlled by a Java (registered trademark) program. There is a form called. In authoring, video and audio are encoded, and the encoded stream and graphics data indicating subtitles, menus, etc. are multiplexed into a transport stream in M2TS format, and playback of playlists, EP maps, etc. is performed. Generate management information necessary for control. Then, the data generated by the authoring is stored in the BD.

  The BD player refers to the management information, separates and decodes the video and audio elementary streams necessary for reproduction, and the graphics data, and outputs them. Here, the video and graphics such as subtitles and menus are output after synthesizing each other's planes. If the resolution of the video and the resolution of the graphics are different, the graphics are combined with the video after up-converting the graphics according to the resolution of the video.

[1-5. Device configuration]]
When playing back content (video) compatible with HDR, a display such as a TV receives and displays an output signal from a playback device such as a BD player. Hereinafter, the display of the video corresponding to the HDR is referred to as “HDR display”, and the display of the video corresponding to the SDR is referred to as “SDR display”. At this time, if the display is compatible with HDR display, the output signal output from the playback device may also be the HDR signal compatible with HDR. On the other hand, when the display does not support HDR display, the playback device converts the output signal into an SDR signal corresponding to SDR and outputs it. The case where the display does not support HDR display means that the display supports only SDR display.

  FIGS. 4A and 4B show an example in which the BD player 200 and the TVs 300 and 310 are connected by HDMI (registered trademark), FIG. 4A shows a case where the TV 300 supports HDR display, and FIG. 4B shows the TV 310. Indicates the case where the display does not correspond to the HDR display. Although the configuration differs between the BD player 200 in FIG. 4A and the BD player 200 in FIG. 4B, FIG. 4A is a diagram showing the case where remapping described later is not performed, and a conversion device that performs remapping The configuration of 210 is omitted.

  In FIG. 4A, BD player 200 reads and decodes video and graphics from media 100. Then, the BD player 200 combines the decoded video and graphics HDR data, and outputs an HDR signal generated by combining to the TV 300 compatible with HDR display by HDMI (registered trademark).

  On the other hand, in FIG. 4B, since the TV 310 does not support HDR display, the BD player 200 uses the HDR EOTF and the SDR EOTF to synthesize the video and graphics HDR data before combining the video and graphics. Remap each to SDR data. Then, the BD player 200 combines the remapped video and graphics SDR data, and outputs an SDR signal generated by combining to the TV 310 that is not compatible with HDR display according to HDMI (registered trademark).

  Note that remapping is processing for converting the first code value in the first EOTF into the second code value in the second EOTF when there are two types of EOTFs, the first EOTF and the second EOTF. In the case of FIG. 4B, the remapping is a process of converting the code value of the HDR EOTF into the code value of the SDR EOTF in the conversion from HDR to SDR.

  That is, in the case of FIG. 4B, the BD player 200 uses the acquisition unit for acquiring the first luminance signal (HDR signal) corresponding to the first luminance range (HDR), the HDR EOTF, and the SDR EOTF. From the code value indicated by the first luminance signal acquired by the acquisition unit, the code value associated by quantization for the second luminance range (SDR) is determined as a converted code value, and the first luminance signal is converted And a converter 210 including a converter for converting the code value into a second luminance signal. More specifically, the conversion unit uses the first EOTF and the second EOTF in the conversion to the second luminance signal to generate a second code corresponding to the code value indicated by the first luminance signal acquired by the acquisition unit. Determine code value as converted code value. Note that the BD player 200 performs a conversion method for performing steps corresponding to each part of the conversion device 210. Although FIG. 4B exemplifies a case where the conversion device 210 converts the HDR signal into an SDR signal and outputs it, as described later, the SDR signal may be converted into an HDR signal and output.

  Since SDR can not express luminance exceeding 100 nit, in the conversion process from HDR to SDR performed by the conversion device 210, at least correspondence between luminance exceeding 100 nit in the HDR signal and the code value of SDR corresponding to the luminance. It is necessary to perform the addition based on a predefined conversion table or an adaptive process according to the luminance distribution of the image in the content. Further, in the conversion process, it is assumed that different conversion rules are required for data such as subtitles in which the luminance value is discrete and video. In addition, since remapping occurs on a frame basis, the processing amount is large particularly for high resolution images such as 4K. Furthermore, since the luminance value changes before and after remapping, the image after remapping may be an image of an impression different from the intention of the producer.

  When converting an HDR signal of video (content) corresponding to HDR into an SDR signal and outputting it, the same remapping as that for video is required for graphics. By remapping both video and graphics, there is also a problem in that the amount of processing required for the remapping is increased, and the image may be converted to a luminance value not intended by the producer.

[1-6. First remap with fixed luminance value]
The luminance range of the graphics master is common to the SDR and HDR masters (see below). That is, the graphics master is generated below the upper limit value of the brightness range of SDR. This is because it is the video such as the main part of the movie that the effect of HDR is the largest in the video content, and the graphics such as subtitles are considered to be less effective than the video.

  In such remapping of graphics, the first remapping is performed to determine the corresponding luminance value from the code value in the first EOTF before conversion, and to determine the code value in the second EOTF after conversion corresponding to the luminance value. . That is, the first remap is a remap with a fixed luminance value in which the code value in the second EOTF is determined using the luminance value determined in the first EOTF as it is.

  Here, a table indicating the correspondence between each of the plurality of code values in each EOTF and each of the plurality of luminance values is held in advance. By referring to each table, a predetermined luminance value corresponding to a predetermined code value is determined, or conversely, a predetermined code value corresponding to a predetermined luminance value is determined.

  Specific processing will be described with reference to FIGS. 5A and 5B. FIG. 5A is an example of the first remapping of the SDR signal to the HDR signal, and is a diagram for describing the first remapping of the SDR signal in which the code value in the SDR is code_sdr. In this first remap, the luminance value val_i corresponding to code_sdr is determined using the SDR EOTF, and then the code_hdr in the HDR EOTF corresponding to the determined luminance value val_i is determined. By this operation, the code value code_sdr in the SDR EOTF is remapped to the code value code_hdr in the HDR EOTF.

  Subsequently, FIG. 5B is an example of the first remapping of the HDR signal to the SDR signal, and in the same manner as in FIG. 5A, a diagram for explaining the remapping of the HDR signal that is the code value code_hdr in the EOTF of HDR. is there. In this first remap, code_hdr is remapped to the code value code_sdr in the SDR EOTF. That is, the luminance value val_i corresponding to code_hdr is determined using the EOTF of HDR, and then the code_sdr in the EDRF of SDR corresponding to the determined luminance value val_i is determined.

  From these facts, the first remap shown in FIGS. 5A and 5B uses (i) the first EOTF to determine the luminance value related to the code value indicated by the first luminance signal, and (ii 2.) For the determined luminance value, the second code value related in the second EOTF is determined as the converted code value.

  In the first remap, when there is no code value matching the luminance value in the first EOTF before conversion in the second EOTF after conversion, the difference in luminance value among the code values of the second EOTF after conversion Choose the code value for which is the smallest. That is, in the first remap shown in FIGS. 5A and 5B, the plurality of second code values associated in the second EOTF does not have the code value associated with the determined luminance value. Among the code values, the code value associated with the luminance value having the smallest difference from the determined luminance value is determined as the converted code value. In the example of FIG. 5A, if there is no code whose luminance value matches val_i (the luminance value related to code-sdr in the SDR EOTF) among the plurality of code values in the HDR EOTF, the HDR Among the plurality of luminance values associated with the plurality of code values in the EOTF, the code value associated with the luminance value closest to val_i is selected.

  As the bit length of the code value of the SDR signal, an 8-bit signal is generally used, but in the HDR signal, the bit length is expanded to 10 bits or 12 bits to express high peak luminance. It is assumed that However, in a conventional authoring system, an optical disc, a BD player or the like, since video and graphics signals are 8 bits, it is desirable to use 8-bit signals from the viewpoint of compatibility. Video is the most important element in content, and the resolution extension from 2K to 4K, ITU-R Recommendation BT. 709 to BT. It is difficult to achieve compatibility with the past because the color gamut is expected to expand to 2020. On the other hand, with regard to graphics, it is also possible to upconvert to 4K when displaying, using conventional 2K, 8 bit, SDR graphics, and combine with video for display. By making the graphics the same as before, there is an advantage that graphics data can be shared at the time of authoring between 2K video and video (content) compatible with SDR and new content video compatible with 4K or HDR. is there.

  Here, in the case where the display corresponds to only the SDR and the SDR EOTF is used, there is no problem because the peak brightness of the SDR can be expressed by 8 bits. On the other hand, when the display supports HDR and HDR EOTF is used, the code length is expanded from 8 bits to 10 bits or 12 bits in HDR EOTF, resulting in the code value The luminance value in the HDR EOTF corresponding to “255”, which is the maximum value in the 8-bit representation, may be smaller than the SDR peak luminance. That is, there is a possibility that the peak luminance of the SDR can not be expressed by corresponding to the 8-bit HDR signal by extracting the values of 0 to 255 among the code values of the EOTF of HDR.

  Therefore, when expressing the HDR EOTF to which 10-bit code values are related by 8-bit code values, the upper 8-bit code values in the HDR EOTF 10-bit code values may be used. Good. Specifically, when converting an 8-bit code value of an HDR signal into a luminance value, the 8-bit HDR signal is shifted up by 2 bits, and a zero is inserted in the lower 2 bits. A code value of bits is generated and a luminance value corresponding to the generated code value is determined.

  That is, in the first remap, when the code value indicated by the acquired first luminance signal is the second bit number smaller than the first bit number represented by the first code value related by the first EOTF, the first remapping is performed. The luminance value associated with the first EOTF is determined by using the upper bits of the second bit number of the code value. Further, in the first remap, the bit length of the first luminance signal is converted to the bit length of the first EOTF, and in the first EOTF, the luminance value corresponding to the code value of the converted first luminance signal is determined.

  The code value generated by shifting up by 2 bits is only a multiple of 4, but the peak luminance of the SDR can be expressed by an 8-bit code value. Alternatively, if the peak brightness of SDR can be expressed by a 9-bit code value, the 9-bit code value may be shifted up by 1 bit, in which case the shifted up code value is a multiple of 2 Take a code value. The same method can be used when the code value of the HDR EOTF is 12 bits.

  As described above, the first remap is a remap in which the luminance value is fixed, and the correspondence between the luminance values between the first and second EOTFs before and after conversion is unnecessary, so the amount of processing related to the remap is reduced. can do. Such a first remap will be referred to as a fixed luminance value remap.

[1-7. Second remap with variable luminance value]
The graphics are described to remap from HDR to SDR or from SDR to HDR by the luminance value fixed remap (first remap). On the other hand, as shown in FIG. 6, in the video, since a master having different peak luminances is used for each of SDR and HDR, the HDR master includes luminance exceeding the peak luminance of SDR. FIG. 6 shows the luminance range of the SDR master and the HDR master in video.

  In the video remap, since the peak luminance of the video HDR signal is higher than the peak luminance of the SDR, the luminance value can not be made constant before and after conversion as in the graphics remap. Therefore, in video remapping, remapping (second remapping) is performed to convert the luminance value before and after conversion. That is, in the second remap, after (i) using the first EOTF, after determining the first luminance value (the luminance value before remapping) related to the code value indicated by the first luminance signal, the first remap is performed. And (ii) determine a second brightness value (brightness value after remapping) in the second brightness range, which is previously associated with the determined first brightness value. Then, in the second remap, similarly to the first remap, the second code value related in the second EOTF is determined as the converted code value for the determined second luminance value.

  FIG. 7 is a diagram for describing an example of relationship information that is a correspondence between the luminance value of HDR and the luminance value of SDR when converting an HDR signal into an SDR signal. Although the details of the mapping method are omitted, in the above correspondence relationship, mapping is performed to the luminance value in the luminance range of SDR so as to keep the luminance value in the low luminance region among the luminance range of HDR as much as possible. The luminance value of the high luminance area, which is a luminance area higher than the low luminance area, is mapped near the peak luminance in the luminance range of the SDR. That is, as shown in FIG. 7, in the determination of the second luminance value, when the determined first luminance value is in the low luminance region on the side where the luminance is low in the first luminance range, the first luminance value is approximately equal to the first luminance value. The second luminance values are determined to be equal luminance values. In the determination of the second luminance value, when the determined first luminance value is in the high luminance region on the high luminance side of the first luminance range, the amount of increase decreases as the first luminance value increases. The second luminance value is determined, and if the determined first luminance value is the maximum luminance value of the first luminance range, the maximum luminance value of the second luminance range is determined as the second luminance value.

  There is also a method of equalizing the luminance value exceeding the SDR peak luminance to the SDR peak luminance by clipping uniformly. That is, in the determination of the second luminance value that is the luminance value after remapping, when the determined first luminance value exceeds the maximum luminance value of the second luminance range, the maximum luminance value of the second luminance range is It may be a luminance value. However, such a method has a disadvantage that it can not express the luminance difference in the high luminance area of the HDR signal at all. In addition, the correspondence of the luminance values of HDR and SDR is the same also when converting from SDR to HDR. A separate table is prepared for the correspondence of the luminance values between HDR and SDR.

  As described above, the second remap in the case where the luminance value changes before and after the remapping is referred to as a luminance value variable remap.

[1-8. Conversion method and conversion device]
FIG. 8 is a block diagram showing a configuration of a remapping unit in the conversion device.

  The remap processing unit 220 is included in the conversion device 210. As shown in FIG. 8, the remap processing unit 220 temporarily stores an EOTF determination unit 221, a processing target determination unit 222, a variable luminance value remapping unit 223, a fixed luminance value remapping unit 224, and a stream of content (video). And a storage unit 225 for storing the data.

  The EOTF determination unit 221 corresponds to an EOTF to which signals of content (video and graphics) read from the medium 100 correspond, and an EOTF to which an output signal to be output to a display such as the TV 300 or 310 displaying an image corresponds. It is determined whether or not. Here, the EOTF to which the output signal corresponds is an EOTF of an output signal that can be displayed by a display such as a TV.

  The processing target determination unit 222 determines whether the processing target is a video (graphics).

  The luminance value variable remapping unit 223 converts the signal of the stream stored in the storage unit 225 into a signal corresponding to the EOTF of the output signal by the luminance value variable remapping (second remapping).

  The luminance value fixed remapping unit 224 converts the stream signal stored in the storage unit 225 into a signal corresponding to the EOTF of the output signal by the luminance value fixed remap (first remap).

  FIG. 9 is a diagram showing a flowchart of remapping processing in the conversion device.

  As shown in FIG. 9, in the remapping process, first, the EOTF determination unit 221 corresponds to the EOTF to which the acquired content (video and graphics) signals correspond, and the output signal to be output to the display. It is determined whether the EOTF is different (step 101). If it is determined "Yes" in step 101, the processes of steps 102 to 104 are performed to convert the method of the luminance range to which the content signal corresponds to the EOTF to which the output signal corresponds. On the other hand, if it is determined "No" in step 101, the remapping process is ended, and the content signal is output without remapping. Thus, by performing step 101, the format of the output signal is determined based on whether the display such as a TV that displays the video is compatible with HDR display. The format of the output signal may be determined to match the main video such as the main story.

  Next, the process target determination unit 222 determines whether the process target is a video (graphics) (step 102). If it is determined "Yes" in step 102, the luminance value variable remapping unit 223 converts the content signal into a signal corresponding to the EOTF of the output signal by the luminance value variable remap (second remap) (step 103).

  On the other hand, if it is determined "No" in step 102, the luminance value fixed remapping unit 224 converts it into a signal corresponding to the EOTF of the output signal by the luminance value fixed remap (first remap) (step 104).

  Thus, when the content (video) is video, the second remapping is performed, and when the content (video) is graphics, the first remapping is performed.

  In the brightness value variable remap of step 103 and the brightness value fixed remap of step 104, a table indicating the correspondence between the brightness values of HDR and SDR is prepared in advance. In this table, in the HDR EOTF and the SDR EOTF, correspondences between luminance values at which code values exist may be described. By doing this, since there is always a code value corresponding to the luminance value of the EOTF after remapping, if there is no code value corresponding to the luminance value, the code value having the luminance value closest to the luminance value is selected. There is no need to search.

  Also, in the luminance value variable remap (second remap), a plurality of tables are adaptively switched based on the luminance distribution within an image or for each scene, or an optimum table is sequentially created for each content, etc. You may That is, for example, in the determination of the second luminance value, which is the luminance value after remapping, a plurality of relationship information (tables) indicating the relationship between the luminance value in the first luminance range and the luminance value in the second luminance expression The relationship information may be selected according to the scene, and the second brightness value may be determined from the determined first brightness value using the selected relationship information.

  In the luminance value variable remap of step 103, processing is performed in the following procedure. In this case, it is assumed that the first luminance signal corresponding to the first EOTF is converted to the second luminance signal corresponding to the second EOTF.

  (1) A first luminance value (luminance value before remapping) corresponding to the code value of the first EOTF is determined.

  (2) A second luminance value (luminance value after remapping) of the second EOTF corresponding to the first luminance value determined in (1) is determined.

  (3) A code value of a second EOTF corresponding to the second luminance value determined in (2) is determined.

  In the luminance value fixed remapping of step 104, processing is performed in the following procedure. In this case, it is assumed that the first luminance signal corresponding to the first EOTF is converted to the second luminance signal corresponding to the second EOTF.

  (1) Determine the luminance value corresponding to the code value of the first EOTF.

  (2) A code value of the second EOTF corresponding to the luminance value determined in (1) is determined.

  * In the case of the fixed luminance value remap, since the luminance value does not change before and after the remapping, the process (2) in step 103 is unnecessary.

  After completion of the remapping process of step 103 and step 104, the conversion device 210 synthesizes and outputs video and graphics. That is, the conversion device 210 may further combine the video and graphics converted into the second luminance signal and output the result by performing the first remapping and the second remapping.

  Furthermore, when outputting to the display by an interface such as HDMI (registered trademark), the conversion device 210 may transmit information for identifying the EOTF of the output signal as meta information. That is, the conversion device 210 may further output the second luminance signal converted from the acquired first luminance signal together with the meta information for identifying the second EOTF.

[1-9. Effect etc]
In the first embodiment, in the reproduction of content, it is determined whether to output in HDR or SDR depending on whether the output destination of the image is HDR compatible, and according to the output format, Remap graphics from SDR to HDR or from HDR to SDR. The fixed luminance remap process in which the luminance value does not change before and after the remap is applied to the graphics, and the variable luminance remap process in which the luminance value may change before and after the remap is applied to the image.

  As for graphics, the brightness does not change before and after remapping, so the image quality intended by the producer can be maintained. In addition, it is not necessary to associate luminance values between before and after conversion EOTF, and the amount of processing related to remapping can be reduced.

Second Embodiment
[2-1. How to generate content]
As shown in Figure 2, in the creation of a video or graphics master, the brightness or hue of each pixel is corrected with respect to a digital image taken with a camera or a scanned image of a film so as to reflect the intention of the producer. In addition to the process of grading, which requires advanced know-how for grading, the required number of man-hours is also enormous. Therefore, it is desirable to minimize the number of masters generated. On the other hand, since HDR and SDR have different peak luminances, it is generally necessary to generate different masters for each. FIG. 10 is a diagram showing an example of combination of HDR and SDR in the case where one content of each of video and graphics streams is included in content. In this example, there are four combinations, which require the HDR and SDR masters for video and graphics, respectively.

  On the other hand, it is the video of the main part of the movie or the like that the effect of HDR is the largest in the video content, and graphics such as subtitles are considered to be less effective than the video. Nevertheless, for graphics as well, creating masters of HDR and SDR as well as video had the problem that the load of content production was heavy.

  Therefore, in the graphics master generation of the present disclosure, as shown in FIG. 11, the SDR and HDR masters are made common. FIG. 11 illustrates generating a graphics master using a common EOTF with a video master. To this end, the luminance range at the graphics master is matched to the SDR luminance range. That is, the peak luminance in the graphics master is equal to or less than the upper limit value of the luminance range of SDR. When mapping graphics data in content to an SDR signal associated with SDR, the code value for each pixel is determined based on the EDR of SDR, and when mapping to an HDR signal associated with HDR , The code value for each pixel is determined based on the EOTF of HDR.

  FIG. 12A is a diagram for describing a case of mapping to an SDR signal in generation of a graphics master. In this case, since the luminance range of the SDR matches the luminance range of the graphics master, all the domain of code values in the SDR EOTF are valid.

  FIG. 12B is a diagram for describing a case of mapping to an HDR signal in generation of a graphics master. In this case, only code values less than or equal to the code value corresponding to the peak luminance of SDR are valid.

  Note that, when the graphics master is mapped to the HDR signal, identification information indicating that the peak luminance is within the luminance range of SDR may be stored in management information such as an elementary stream or a playlist. . In the remapping process described above, it is possible to determine which of the fixed luminance value remap and the variable luminance value remap is to be applied based on the identification information. In addition, when outputting with an interface such as HDMI (registered trademark), the identification information may be stored as meta information of the output interface.

  Graphics EOTF can be determined to match the video. That is, if the video is HDR, the graphics data is also HDR, and if the video is SDR, the graphics data is also SDR. Alternatively, graphics data may always be SDR.

  The same idea can be applied when there are multiple videos. For example, if there is a sub video to be superimposed or displayed side by side with the main video, the sub video EOTF can be adjusted to the main video, and so on.

[2-2. Data generation method and apparatus]
FIG. 13 is a block diagram showing a configuration of a generation unit that generates a graphics signal in authoring.

  The generation unit 400 includes a GFX grading unit 410, a determination unit 420, an HDR signal generation unit 430, and an SDR signal generation unit 440.

  The GFX grading unit 410 grades the graphics master so that the luminance value is equal to or less than the peak luminance of SDR.

  The determination unit 420 determines whether the video displayed simultaneously with the graphics is HDR.

  When the determination unit 420 determines that the video displayed simultaneously with the graphics is the HDR, the HDR signal generation unit 430 converts the luminance value of the graphics into a code value using the EOTF of the HDR.

  When the determination unit 420 determines that the video displayed simultaneously with the graphics is not HDR (that is, SDR), the SDR signal generation unit 440 converts the luminance value of the graphics into a code value using the SDR EOTF. Do.

  FIG. 14 is a flowchart showing a method of generating a graphics signal in authoring.

  First, the GFX grading unit 410 grades the graphics master so that the luminance value is equal to or less than the peak luminance of SDR (step 201).

  Next, the determination unit 420 determines whether the video displayed simultaneously with the graphics is HDR (step 202).

  If the HDR signal generation unit 430 determines “Yes” in step 202, the HDR signal generation unit 430 converts the luminance value of graphics into a code value using the EOTF of HDR (step 203).

  When the SDR signal generation unit 440 determines “No” in step 202, the SDR signal generation unit 440 converts the luminance value of graphics into a code value using the EDR of the SDR (step 204).

  Although it is determined in step 202 whether the graphics are displayed simultaneously with the video, for example, if the graphics are subtitles, the video on which the subtitles are superimposed is determined. Also, for graphics that are not displayed simultaneously with the video, such as a menu, it may be determined based on whether or not the main video is HDR. In addition, in order to use the same format as the conventional 2K format for graphics, the process of step 204 may always be performed without performing the determination process of step 202, assuming that conversion is always performed using SDR EOTF. .

  As described above, by setting the luminance range of graphics below the peak luminance of SDR, there is an advantage that it is possible to perform fixed remapping of the luminance value without changing the luminance value before and after remapping.

  In addition, grading so that the luminance of the HDR master falls within the range of SDR is also possible for data other than graphics. Furthermore, particularly in graphics such as subtitles, the merit of using a luminance value higher than the peak luminance of SDR is small. Therefore, in the remapping from SDR to HDR, the luminance value fixed remapping may be applied regardless of whether the grading is within the range of SDR.

[2-3. Effect etc]
In the generating apparatus and method according to the second embodiment, when authoring content that includes video data such as graphics in addition to video, for video data other than video, a master common to HDR and SDR is used. use. Therefore, the peak luminance in the master is graded to be within the luminance range of SDR.

  Thereby, since masters other than video can be made common to HDR and SDR, the man-hours concerning master generation can be reduced.

(Other embodiments)
As described above, the embodiment has been described as an example of the technology disclosed in the present application. However, the technology in the present disclosure is not limited to this, and is also applicable to embodiments in which changes, replacements, additions, omissions, and the like are appropriately made. Moreover, it is also possible to combine each component demonstrated by the said embodiment, and to set it as a new embodiment.

  So, below, other embodiments are illustrated.

  For example, in each of the above-described embodiments, two types of HDR and SDR have been described as the format of the output signal output from the conversion device 210. When outputting to HDMI (registered trademark) or the like, it is output by either HDR or SDR as a standard. For example, when a BD player is built in a TV, or a broadcast is received and reproduced by a TV Alternatively, when viewing the OTT service on a tablet or the like, it is possible to output a signal directly from the conversion device 210 to the display device.

  At this time, when the peak luminance in the HDR standard and the peak luminance that can be displayed on the display device are different, remapping processing according to the EOTF of the display device is performed on data in the content corresponding to the HDR. Good. In addition, as to the SDR or HDR signal input to the display device as HDMI (registered trademark), remapping processing may be performed again on the EOTF according to the peak luminance of the display device.

  That is, in this case, the conversion to the second luminance signal sets the acquired first luminance signal as the first EOTF and the luminance range that can be displayed on the display device to which the second luminance signal is output. It may be performed using the second EOTF.

  In each of the above embodiments, although not mentioned, in BD authoring, it is possible to specify video, audio, or graphics to be reproduced in units of play items in the playlist. As described above, when video or audio to be reproduced in units of play items, or graphics are designated, in the interface such as HDMI (registered trademark), when HDR and SDR are switched in units of play items, Reset processing may be applied at the boundary, and seamless playback may not be possible. Therefore, when HDR and SDR are switched between play items connected seamlessly, remapping processing is performed in the conversion device provided in the BD player or the like so that the EOTF of the output signal becomes the same as the previous play item. You may go. Alternatively, switching of the EOTF may be prohibited between play items connected seamlessly, and identification information indicating that the EOTF is not switched may be stored in management information such as a playlist.

  In addition, the authoring or conversion method of each of the above-described embodiments can be applied not only to packaged media such as optical disks but also to broadcasting and Over The Top (OTT) services. For example, in broadcasting, in addition to the main program of a broadcast program, data broadcasting transmitted by broadcasting or content acquired via a communication network can be superimposed and displayed on the video of the main program. At this time, it is expected that the main video will be a mixture of the HDR program and the SDR program, and the peak luminance limitation according to the method described so far for graphics and video in contents acquired separately from the main program. And remapping can be performed.

  In each of the above embodiments, EOTFs having different peak luminances such as HDR and SDR have been described, but the same concept can be applied to color gamut and bit depth. As for the color gamut, as the resolution is expanded from 2K to 4K, BT. 709 color space from BT. Change to color space to 2020. FIG. 15 shows the BT. 709 and BT. FIG. 10 is a diagram showing a C. 709 color space in a CIE color system; In 2020, BT. It can be seen that the color gamut is expanded compared to 709. As in the case of matching the peak luminance of the HDR signal to the peak luminance of SDR in the graphics, BT. When using the 2020 color space, BT. Colors within the color gamut of 709 can be used. As a color space, BT. When using 2020, the color is BT. Similarly to the EOTF, whether or not it is within the range of 709 can be stored in the content or as meta information of the output interface.

  In the above embodiments, each component may be configured by dedicated hardware or may be realized by executing a software program suitable for each component. Each component may be realized by a program execution unit such as a CPU or a processor reading and executing a software program recorded in a recording medium such as a hard disk or a semiconductor memory.

  As mentioned above, although the conversion method and conversion device concerning one or a plurality of modes of this indication were explained based on an embodiment, this indication is not limited to this embodiment. Without departing from the spirit of the present disclosure, various modifications that may occur to those skilled in the art are applied to the present embodiment, and a form constructed by combining components in different embodiments is one or more of the present disclosure. It may be included within the scope of the embodiments.

  The present disclosure is useful as a conversion method, conversion apparatus, and the like that can appropriately convert luminance from the first luminance range to the second luminance range in which the luminance range is expanded or reduced.

100 media 200 BD player 210 conversion device 220 remap processing unit 221 determination unit 222 processing object determination unit 223 luminance value variable remapping unit 224 luminance value fixed remapping unit 225 storage unit 300, 310 TV
400 generation unit 410 grading unit 420 determination unit 430 HDR signal generation unit 440 SDR signal generation unit

Claims (9)

A brightness conversion method for converting the brightness of an image, comprising:
Acquiring a first luminance signal that is a luminance signal of the image;
Converting a first code value included in the first brightness signal into a first brightness value using a first EOTF (Electro-Optical Transfer Function) indicating a correspondence between the brightness value of the first brightness range and the code value;
Converting the first luminance value into a second luminance value previously associated with the first luminance value;
Converting the second brightness value into a second code value using a second EOTF indicating a correspondence between a brightness value of a second brightness range different from the first brightness range and a maximum value and a code value;
Outputting a second luminance signal including the second code value;
Brightness conversion method. When the first luminance value is in a low luminance area of the first luminance range, the first luminance value is converted into the second luminance value having a luminance value substantially equal to the first luminance value,
When the first brightness value is in a high brightness area of the first brightness range, the second brightness is such that the increase amount of the brightness value decreases as the first brightness value increases. Convert to a value,
The luminance conversion method according to claim 1. When the first luminance value exceeds the maximum luminance value of the second luminance range, the first luminance value is converted into the second luminance value which is the maximum luminance value of the second luminance range.
The luminance conversion method according to claim 1. Determining a conversion method from the first luminance value to the second luminance value according to a scene of the video;
The luminance conversion method according to claim 1. A luminance conversion device that performs luminance conversion of an image,
An acquisition unit configured to acquire a first luminance signal that is a luminance signal of the image;
(I) A first code value included in the first brightness signal is converted to a first brightness value using a first EOTF (Electro-Optical Transfer Function) indicating a correspondence between the brightness value of the first brightness range and the code value. Converting (ii) converting the first luminance value into a second luminance value previously associated with the first luminance value; and (iii) converting the second luminance value into the first luminance range and the maximum value. A converter for converting into a second code value using a second EOTF indicating the correspondence between the luminance value of the second luminance range and the code value in different
An output unit that outputs a second luminance signal including the second code value;
A luminance conversion device having: The conversion unit is
When the first luminance value is in a low luminance area of the first luminance range, the first luminance value is converted into the second luminance value having a luminance value substantially equal to the first luminance value,
When the first brightness value is in a high brightness area of the first brightness range, the second brightness is such that the increase amount of the brightness value decreases as the first brightness value increases. Convert to a value,
The luminance conversion device according to claim 5. The conversion unit is
When the first luminance value exceeds the maximum luminance value of the second luminance range, the first luminance value is converted into the second luminance value which is the maximum luminance value of the second luminance range.
The luminance conversion device according to claim 5. The conversion unit is
Determining a conversion method from the first luminance value to the second luminance value according to a scene of the video;
The luminance conversion device according to claim 5. A luminance conversion device according to claim 5;
A display device for displaying the image based on the second luminance signal;
A video display device having

Images