JP2018117281A - Character superimposition synthesizer and program thereof - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a character superimposition synthesizer that can make it easy to see superimposed characters.SOLUTION: A character superimposition synthesizer 1 includes a video detector 11 for detecting a determination signal from an input video, a video determination unit 12 for determining whether the input video is HDR system or SDR system based on the determination signal, a video level determination unit 13 for determining the video level of superimposed characters, based on the determination results from the video determination unit 12, a character superimposition video level instruction unit 14 for instructing a character superimposition presentation unit 16 to present superimposed characters according to the video level, a character superimposition generation unit 15 for generating superimposed characters to be synthesized with the input video, a character superimposition presentation unit 16 for presenting superimposed characters with a video level determined by the video level determination unit 13, and a character superimposition synthesis unit 17 for synthesizing superimposed characters with the input image.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a character super synthesizer and a program thereof.

  In current television broadcasting, characters are superimposed on the video to convey information in an auxiliary manner. Hereinafter, the superimpose is referred to as “super”, and the character superposed on the video is referred to as “character super”.

Conventional television broadcasts are expressed in a standard dynamic range (SDR) system with a certain contrast between light and dark in a television receiver such as a CRT (Cathode Ray Tube) or a liquid crystal television. Also, in television video production in a broadcasting station, when producing video in a video editing studio or the like, generally, the brightness of the monitor background in the viewing environment is in accordance with ITU-R recommendation BT.2035. , An environment of 10 (8-12) cd / m ² corresponding to 10% ± 2% of the standard white level (maximum brightness) of the image shown on the monitor. As a reference for the brightness of the production monitor, the standard white level 100% of the relative video level (IRE value) from 0% to 100% (or 109%) is set to the brightness 100 cd / m ^2. Yes. That is, the broadcast video is produced at a relative video level of 0% to 100%. And when it is broadcasted to the home, depending on the luminance characteristics that can be expressed on the television receiver side, the viewing environment (generally brighter than the studio production environment), the viewer's preference, the maximum luminance is approximately 1 to In many cases, viewing is performed by adjusting values such as contrast and brightness to about 4 times (400 cd / m ² ). At this time, in the case of an achromatic color, the character supermarket is generally broadcast with the maximum luminance video level set to a standard white level of 100%.

In recent years, with the background of technological advancements such as improving the maximum luminance performance of displays such as television receivers and improving the brightness / darkness performance that can be obtained with an image sensor, the difference in brightness has been expanded to broaden the range of expression, making it a more photographic subject. Video display technology based on a high dynamic range (HDR) method that faithfully represents (scene) is being applied. The maximum luminance of the display is becoming able to express 700 cd / m ² or more on a liquid crystal monitor (contrast ratio 1000: 1 or more). In a display using an organic EL element (OLED) which is a self-luminous display device that has recently emerged, the contrast ratio exceeds 1,000,000: 1, and the maximum brightness is also 500 cd / m ² to 1000 cd / m ^2, with a light / dark width. The range that can be expressed is expanding.

  The conventional opto-electric transfer function (OETF) in the SDR system is the video level (electric signal, relative value) E, and the level (relative value) obtained by converting the scene luminance into an electric signal. When L, it is represented by the following formula (1). In Equation (1), the gamma γ value is 2.2 or 2.4 (rounded to the second decimal place).

  Further, in one of the HDR systems, in the hybrid log-gamma (HLG) system, OETF is, for example, the following formula (2) when the video level (electrical signal, relative value) is E ′. (Refer to Non-Patent Documents 1 and 2).

  Here, in the formula (2), the above formula is the formula (2-a) and the lower formula is the formula (2-b). r is the video level of the boundary point between the power function of Expression (2-a) and the function of Expression (2-b), and represents the reference white level. The constants a, b, and c are constants defined in advance by the HLG method. The graph of Formula (2) is shown in FIG. In FIG. 6, the vertical axis represents the video level E and the horizontal axis represents the scene luminance L. For example, when the scene luminance at the boundary point is L = 1, the video level at the boundary point is r = E ′ according to the equation (2-a). As shown in FIG. 6, the equations (2-a) and (2-b) must have continuity from the boundary point. Note that the video level (reference white level) r at the boundary point may be another value such as 0.63, 0.7, or 0.75 in addition to 0.5. At this time, the values of constants a, b, and c in equation (2) may change so as to have continuity.

  Since the power function of Expression (2-a) is similar to Expression (1), when the relative video level is 50% or less, it is possible to perform video expression with high compatibility between the SDR method and the HDR method. Further, the function of the equation (2-b) represents that a bright part (highlight) of an image can be expressed when the relative video level exceeds 50%.

  For this reason, in television broadcasting, when transitioning from conventional SDR video broadcasting to HDR video broadcasting, when the HLG method is applied, a relative video level of 0 to 50% is an expression area compatible with SDR video. The relative video level 0 to 100% obtained by combining this area and the highlight area can be used as the HDR video expression area.

  Therefore, when a broadcast program produced by the conventional SDR method is to be expressed as a faithful broadcast by an HDR method, the image recorded by the HDR method is expressed by a relative video level of 0 to 100%. It is assumed that It is assumed that the video recorded in the SDR system is expressed at a relative video level of 0 to 50 (or 63, 70, 75)%. For images expressed at such relative video levels of 0 to 100%, the character super has been superposed so far at a maximum level of 100% (or a level that is at most 20% lower than that). Here, in the case of superimposing characters on a broadcast program produced by the SDR system expressed by the relative video level 0 to 50 (or 63, 70, 75)%, the relative video level corresponding to the maximum luminance of the SDR system. It will be expressed in 50 (or 63, 70, 75)%.

Association of Radio Industries and Business (ARIB) Standard STD-B67 Ver1.0 International Telecommunication Union Radiocommunications Sector (ITU-R) Recommendation BT.2100-0 (07/2016)

  However, when HDR video and SDR video coexist, if the character super is expressed at a relative video level of 50%, the character super will be expressed in a neutral white as in the conventional SDR method, but the character super is dark. The problem that it becomes difficult to see arises. Furthermore, since the video levels in the HDR method and the SDR method are different, when a character super is synthesized (superimposed) at a certain video level, the character super is far from the brightness of the video, and the viewer feels uncomfortable. Also occurs.

  Accordingly, it is an object of the present invention to provide a character super composition device and a program for making it easy to see a character super image even when HDR video and SDR video are mixed.

  In view of the above-described problems, a character super synthesizer according to the present invention is a character super synthesizer that synthesizes a character super image with an HDR or SDR input image, and includes a video detection unit, a video determination unit, and a video. The level determination unit and the character super synthesis unit are provided.

According to such a configuration, the character super synthesizer detects a determination signal for determining whether the input video is the HDR system or the SDR system from the input video by the video detection unit.
In the character super synthesizer, the video determination unit determines whether the input video is the HDR method or the SDR method based on the determination signal detected by the video detection unit.

In addition, when the input video is in the HDR format, the character super synthesizer uses the HDR character super image level preset between 50% and 100% of the relative video level. Decide to present.
Furthermore, when the input video is in the SDR system, the text super synthesizer determines that the text super is presented at the SDR text super video level set in advance based on the maximum level of the SDR system.
Thereafter, the character super synthesizer synthesizes the character super image with the input video at the video level determined by the video level determination unit by the character super synthesis unit.

  As described above, the character super composition device synthesizes the character super with brightness according to the input video such as the HDR method or the SDR method without making the character super image level constant. .

  The character super synthesizer according to the present invention synthesizes the character super with brightness according to the HDR method or the SDR method, so that the character super can be easily seen.

It is a block diagram which shows the structure of the character super-synthesis apparatus which concerns on embodiment of this invention. It is explanatory drawing explaining the synthesis | combination of the character superposition by the character supersynthesizing apparatus of FIG. It is a flowchart which shows operation | movement of the character super composition apparatus of FIG. In the Example of this invention, it is a histogram showing the evaluation result in case the total average luminance level of a HDR evaluation image | video is 54%. In the Example of this invention, it is a histogram showing the evaluation result in case the total average luminance level of a HDR evaluation image | video is 40%. It is a graph which shows OETF of a hybrid log gamma system.

[Character super synthesizer configuration]
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings as appropriate.
With reference to FIG. 1, the structure of the character super composition apparatus 1 which concerns on embodiment of this invention is demonstrated.

  The character super synthesizer 1 synthesizes a character super image with a HDR or SDR input video. In this embodiment, the character super composition apparatus 1 is used for production of a broadcast program.

  As shown in FIG. 1, a character super synthesizer 1 includes a video input unit 10, a video detection unit 11, a video determination unit 12, a video level determination unit 13, a character super video level command unit 14, a character super A generation unit 15, a character super presentation unit 16, and a character super composition unit 17 are provided.

  The video input unit 10 receives input video from the outside (for example, a camera, a recording device, a video editing device, or a video transmission device). This input video is a video in which HDR video and SDR video are mixed. For example, the input video includes HDR video having a dynamic range of HLG, PQ (Perceptual Quantization), etc., and conventional SDR video. The resolution of the video signal is not particularly limited, such as 8K (horizontal 7680 × vertical 4320 pixels), 4K (horizontal 4320 × vertical 2160 pixels), and 2K (horizontal 1920 × vertical 1080 pixels). Then, the video input unit 10 outputs the input video to the video detection unit 11 and the character super composition unit 17.

The video detection unit 11 detects a determination signal for determining whether the input video is in the HDR system or the SDR system from the input video input from the video input unit 10. Then, the video detection unit 11 outputs the detected determination signal to the video determination unit 12.
Details of the video detection unit 11 will be described later.

  The video determination unit 12 determines whether the input video is in the HDR system or the SDR system based on the determination signal input from the video detection unit 11. Here, when the input video is in the HDR format, the video determination unit 12 outputs a video level determination command to the HDR character super video level determination unit 131. On the other hand, when the input video is the SDR system, the video determination unit 12 outputs a video level determination command to the SDR character super video level determination unit 133. Hereinafter, three determination methods will be specifically described.

<First Example: Method for Determining with Auxiliary Data>
This first example is a technique for detecting and determining auxiliary data superimposed on an input video as a determination signal.

  The auxiliary data stores a unique ID (IDentifier) in each of the HLG method, the PQ method, and the SDR method. For example, if the input video is SMPTE292, ITU-R recommendation BT. Consider a case corresponding to a serial digital interface (HD-SDI) of an HDTV signal proposed in 1120, ARIB standard BTA S-004. In this case, the auxiliary data is a unique ID added to the user data word area in the auxiliary signal superposition method proposed in SMPTE291.

  Also, consider the case where the input video is an ultra-high definition video such as 4K or 8K. In this case, the auxiliary data is a unique ID added to an auxiliary signal area such as an interface for transmitting an ultra-high definition video proposed in the ARIB standard STD-B58.

  In this first example, the video detection unit 11 detects auxiliary data from the input video as a determination signal. Then, the video determination unit 12 refers to the auxiliary data detected by the video detection unit 11 and determines whether the input video is in the HDR system (HLG system, PQ system) or the SDR system.

<Second Example: Method for Determining at Mapping Level>
This second example is a technique for detecting a relative video level (IRE value) to which an input video is mapped as a determination signal and performing the determination. Here, it is assumed that the SDR video is presented in a state in which compatibility is maintained on a display set to HLG OETF.

  There may be a case where the auxiliary data as in the first example is not added to the input video. In this case, it is determined whether the HDR system (HLG system, PQ system) or SDR system is used according to the relative video level to which the input video is mapped.

  As shown in Equation (2), the SDR video is mapped at a relative video level of 0% to r% (where r is 50%, 63%, 70%, or 75%). Therefore, the video detection unit 11 reproduces a preset section (for example, 10 seconds) from the beginning of the input video as the determination signal, and detects the maximum value (maximum level) of the relative video level. Then, the video determination unit 12 determines whether the input video is the HDR method or the SDR method based on the threshold determination of the maximum level detected by the video detection unit 11. For example, the video determination unit 12 sets the threshold value to 80% in advance, determines the HDR method when the maximum level is 80% or more, and determines the SDR method when the maximum level is less than 80%.

  Note that a video (for example, picture-in-picture) in which an SDR video is inserted into an HDR video is considered. In this case, the video determination unit 12 determines that the video is an HDR video because the maximum level of the video is the same as that of the HDR system.

<Third example: Method of judging by credit information>
This third example is a technique in which when credit information indicating the video format of the input video is added before the head of the input video, this credit information is detected as a determination signal and the determination is performed. Note that the credit information is a video reference signal such as a color bar including an image area representing the maximum white (relative video level is 100%).

  This third example is a case where the transmission is limited to the HDR method by HLG, and the HDR video is a video represented by a combination of formula (2-a) and formula (2-b) in which the maximum white is 100%. This is a technique that can be applied only when a signal is prepared and a video signal of maximum 50% white represented by the compatible expression (2-b) is prepared in advance and each is input. . The third example is not applied when an SDR image prepared with the conventional maximum white level of 100%, for example, represented by the equation (1) is input.

  Here, the video detection unit 11 detects credit information as a determination signal. Then, the video determination unit 12 determines whether the input video is in the HDR system or the SDR system depending on whether or not the video detection unit 11 has detected the credit information. For example, when the maximum white in the credit information is detected to be 100%, the video determination unit 12 determines that the HDR method is used, and the maximum white in the credit information can be detected as r% (for example, 50%). In this case, the SDR method is determined.

Since the input video is produced in broadcast program units, scene units, or cut units, the HDR system or SDR system is switched in these units. Therefore, in the first to third examples, the HDR method or the SDR method is not frequently switched as in units of frames.
In addition, it is set in advance whether the video detection unit 11 and the video determination unit 12 use the first to third examples.

The description of the configuration of the character super synthesizer 1 will be continued.
The video level determination unit 13 determines the video level of the text super based on the determination result of the video determination unit 12, and includes an HDR character super video level determination unit 131, an SDR character super video level determination unit 133, Is provided.

  The HDR character super video level determination unit 131 determines to present a character super at a predetermined HDR character super video level when a video level determination command is input from the video determination unit 12. Then, the HDR character super video level determination unit 131 outputs the determined video level to the character super video level command unit 14.

  For example, the video producer presets the HDR character super video level with an arbitrary value from 50% (intermediate value) to 100% (maximum value) of the relative video level. At this time, the HDR character super video level is preferably set in advance between 70% and 100% of the relative video level for the reason described later (see the embodiment).

  The SDR character super video level determination unit 133 determines to present a character super at a predetermined SDR character super video level when a video level determination command is input from the video determination unit 12. Then, the SDR character super image level determining unit 133 outputs the determined image level to the character super image level command unit 14.

  For example, the video producer sets the SDR character super video level in advance based on the maximum level of the SDR method. At this time, the character super video level for SDR may be set in advance between the relative video level of the SDR system (the maximum level of the input video) and between 50% and 100% of the relative video level.

  The text super video level command unit 14 instructs the text super presentation unit 16 to present a text super in accordance with the video level input from the video level determination unit 13.

  The character super generation unit 15 generates a character super combined with the input video in accordance with an external (for example, video producer) instruction. Here, the character super generation unit 15 can arbitrarily set the position, font size, color, decoration (shadow, character background), and the like of the character super to be generated. Then, the character super generating unit 15 outputs the generated character super to the character super presenting unit 16.

  The character super presentation unit 16 presents (outputs) the character super input from the character super generation unit 15 at the video level determined by the video level determination unit 13. That is, the character super presenting unit 16 outputs the character super of the video level determined by the video level determining unit 13 to the character super combining unit 17.

The character super combining unit 17 combines the character super input from the character super presenting unit 16 with the input video from the video input unit 10. For example, as shown in FIG. 2, the character supersynthesizing unit 17 synthesizes a character superposition of “weather information: heavy snow warning in XX region” with a news program (input video). At this time, the character supermarket becomes bright according to the video level of the news program.
Thereafter, the character super composition unit 17 outputs the input video in which the character super is composed to the outside (for example, a broadcast transmission device).

[Operation of character super synthesizer]
With reference to FIG. 3, the operation of the character super composition apparatus 1 will be described (see FIG. 1 as appropriate).
As shown in FIG. 3, the video input unit 10 receives an input video from the outside (step S1).
The video detection unit 11 detects a determination signal for determining whether the input video is in the HDR system or the SDR system from the input video input in step S1 (step S2).

The video determination unit 12 determines whether the input video is in the HDR format or the SDR format based on the determination signal detected in step S2 (step S3).
If the input video is in the HDR format (HDR in step S3), the character super synthesizer 1 proceeds to the process in step S4.
If the input video is in the SDR format (SDR in step S3), the character super synthesizer 1 proceeds to the process in step S5.

The video level determination unit 13 (HDR character super video level determination unit 131) determines to present the character super at the HDR character super video level (step S4).
The video level determination unit 13 (SDR character super video level determination unit 133) determines to present the character super at the SDR character super video level (step S5).

The character super image level command unit 14 instructs the character super presenter 16 to present a character super in accordance with the image level determined in step S4 or step S5 (step S6).
The character super generating unit 15 generates a character super combined with the input video (step S7).

The character super presentation unit 16 presents the character super generated in step S7 at the video level determined in step S4 or step S5 (step S8).
The character super synthesizer 17 synthesizes the character super presented in step S8 with the input video input in step S1 (step S9).
As described above, the character super synthesizer 1 synthesizes the character super with the brightness according to the input video such as the HDR method or the SDR method, so that the character super can be easily seen.

As mentioned above, although each embodiment of this invention was explained in full detail, this invention is not limited to above-mentioned each embodiment, The design change etc. of the range which does not deviate from the summary of this invention are included.
The present invention is similarly applied not only to an interface of a broadcast production apparatus but also to superimposition on a broadcast video signal and an auxiliary signal that are broadcast to a home. That is, the present invention can also be applied to a television receiver.

  In the above-described embodiment, the character super synthesizer has been described as independent hardware, but the present invention is not limited to this. For example, hardware resources such as a CPU, a memory, and a hard disk included in a computer can be realized by a character super synthesizer program that operates cooperatively as a character super synthesizer. This program may be distributed via a communication line, or may be distributed by writing in a storage medium such as a CD-ROM or a flash memory.

(Example)
First, a character super was synthesized on the video produced by OETF represented by Formula (2) by the HLG method. At this time, the character super is set step by step with a predetermined IRE value (50%, 60%, 70%, 75%, 80%, 85%, 90%, 100%), and it is a simple color with no achromatic decoration. The character font. The character supermarket was synthesized so that it can be recognized as character information at a position of about 7% to 13% from the lower end of the screen of the evaluation video monitor described later. Hereinafter, this video is referred to as an HDR evaluation video.

In addition, in order to present an HDR evaluation video, an evaluation video monitor in which an HLG EOTF, which is an inverse function of Equation (2), is prepared. At this time, the evaluation video monitor used an organic EL display device having a maximum luminance of 1000 cd / m ² and had a contrast ratio of 1 million to 1 or more.

The viewing environment was an ambient environment with a gray background and a dim light. At this time, the luminance of the background was 5 cd / m ² . The viewing distance was 3H with respect to the height H of the screen. Then, 14 video professionals who have experience in broadcasting technology work as evaluators. When the HDR evaluation video is presented on the evaluation video monitor, the range of the relative video level of the character supermarket that is considered preferable (to one) (Not limited to multiple).

The evaluation results are shown in FIGS. FIG. 4 shows, as a histogram, the ratio of evaluators who answered that they were preferable to the character supermarket presented at each relative video level when the total average luminance level of the HDR evaluation video was 54%. FIG. 5 is a histogram showing the ratio of evaluators who answered that the superimpose presented at each relative video level is preferable when the total average luminance level of the HDR evaluation video is set to 40%. .
In FIGS. 4 and 5, the vertical axis is the rate of evaluators who answered that it is preferable, and the horizontal axis is the relative video level.

  In FIG. 4, the superimpose presented at a relative video level of less than 70% in FIG. 4 and less than 60% in FIG. The reason for this is considered to be that the character supermarket intended to be presented in white characters common in SDR video broadcast programs appeared gray and was not answered as preferable.

  In the SDR system, the maximum level (100%) is generally used as the relative video level when presenting a character super. On the other hand, in the HDR evaluation video, the mode value of the maximum level that the evaluator feels preferable is distributed in a value between the maximum value and the intermediate value (50%). The reason is considered that the maximum level was felt dazzling on the evaluation video monitor.

  Also, as a reference example, an image was prepared in which the background was a single black color with a relative video level of 0% and a character supermarket with a relative video level of 50% was synthesized. Then, when this video was presented on the evaluation video monitor, it was evaluated in the same viewing environment and the same evaluator whether it looked white or gray.

  As for the evaluation result of the reference example, 100% of the evaluators answered that it looks white, whereas the evaluator who answered that it looks gray is 0%. In other words, in this reference example, a character supermarket with a relative video level of 50% answers that all evaluators look white. The reason is considered that when the video is not HDR and the video level is low, the video can be seen as a character super as usual even at a relative video level of 50%. Furthermore, when the background does not exceed the brightness of the character super, the character super is preferably set to 50% or more of the maximum value of the compatible SDR method in the formula (2), and the evaluator does not feel uncomfortable. It is thought that he did not answer.

DESCRIPTION OF SYMBOLS 1 Character super composition apparatus 10 Image | video input part 11 Image | video detection part 12 Image | video determination part 13 Image | video level determination part 14 Character super image level command part 15 Character super production | generation part 16 Character super presentation part 17 Character super composition part

Claims (6)

A character super synthesizer that synthesizes a character super image with an HDR video or SDR video input,
A video detection unit for detecting a determination signal for determining whether the input video is an HDR method or an SDR method from the input video;
A video determination unit that determines whether the input video is an HDR method or an SDR method based on a determination signal detected by the video detection unit;
When the input video is in the HDR format, it is determined that the character super is presented at a preset HDR text super video level between 50% and 100% of the relative video level, and the input video is in the SDR format. A video level determination unit that determines to present the character super at a character super video level for SDR set in advance based on the maximum level of the SDR method;
A character super combining unit that combines the character super image with the input image at the image level determined by the image level determining unit;
A character super synthesizer characterized by comprising:   The character super synthesizer according to claim 1, wherein the video level determination unit presets the HDR character super video level between 70% and 100% of the relative video level.   2. The video level determining unit according to claim 1, wherein the SDR character super video level is preset between the maximum level of the input video and between 50% and 100% of the relative video level. The character supersynthesizing device according to claim 1 or 2. The video detection unit detects the maximum value of the relative video level in a preset section from the beginning of the input video as the determination signal,
4. The method according to claim 1, wherein the video determination unit determines whether the input video is an HDR method or an SDR method based on a threshold determination of a maximum value detected by the video detection unit. The character supersynthesizing device according to one item.   When the input video is an HDR video including an SDR video, the video level determination unit is configured such that the HDR character super video level is preset between 70% and 100% of the relative video level. The character supersynthesizing device according to any one of claims 1 to 4.   A character supersynthesizing program for causing a computer to function as the character supersynthesizing device according to any one of claims 1 to 5.

Images