WO2020007165A1 - 一种视频信号处理的方法及装置 - Google Patents

一种视频信号处理的方法及装置 Download PDF

Info

Publication number
WO2020007165A1
WO2020007165A1 PCT/CN2019/090566 CN2019090566W WO2020007165A1 WO 2020007165 A1 WO2020007165 A1 WO 2020007165A1 CN 2019090566 W CN2019090566 W CN 2019090566W WO 2020007165 A1 WO2020007165 A1 WO 2020007165A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
brightness
linear
display
rgb
Prior art date
Application number
PCT/CN2019/090566
Other languages
English (en)
French (fr)
Inventor
王正
袁乐
吴仁坚
黄芳
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to KR1020217000634A priority Critical patent/KR102511572B1/ko
Priority to JP2020568530A priority patent/JP7115776B2/ja
Priority to CN201980025155.1A priority patent/CN112272947B/zh
Priority to EP19830987.4A priority patent/EP3796648A4/en
Publication of WO2020007165A1 publication Critical patent/WO2020007165A1/zh
Priority to US17/135,568 priority patent/US11317071B2/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N9/00—Details of colour television systems
    • H04N9/64—Circuits for processing colour signals
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00—Details of television systems
    • H04N5/14—Picture signal circuitry for video frequency region
    • H04N5/20—Circuitry for controlling amplitude response
    • H04N5/202—Gamma control
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N9/00—Details of colour television systems
    • H04N9/64—Circuits for processing colour signals
    • H04N9/67—Circuits for processing colour signals for matrixing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N11/00—Colour television systems
    • H04N11/06—Transmission systems characterised by the manner in which the individual colour picture signal components are combined
    • H04N11/20—Conversion of the manner in which the individual colour picture signal components are combined, e.g. conversion of colour television standards
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/90—Dynamic range modification of images or parts thereof
    • G06T5/92—Dynamic range modification of images or parts thereof based on global image properties
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G5/00—Control arrangements or circuits for visual indicators common to cathode-ray tube indicators and other visual indicators
    • G09G5/10—Intensity circuits
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00—Details of television systems
    • H04N5/44—Receiver circuitry for the reception of television signals according to analogue transmission standards
    • H04N5/57—Control of contrast or brightness
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/01—Conversion of standards, e.g. involving analogue television standards or digital television standards processed at pixel level
    • H04N7/0135—Conversion of standards, e.g. involving analogue television standards or digital television standards processed at pixel level involving interpolation processes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N9/00—Details of colour television systems
    • H04N9/64—Circuits for processing colour signals
    • H04N9/68—Circuits for processing colour signals for controlling the amplitude of colour signals, e.g. automatic chroma control circuits
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N9/00—Details of colour television systems
    • H04N9/64—Circuits for processing colour signals
    • H04N9/68—Circuits for processing colour signals for controlling the amplitude of colour signals, e.g. automatic chroma control circuits
    • H04N9/69—Circuits for processing colour signals for controlling the amplitude of colour signals, e.g. automatic chroma control circuits for modifying the colour signals by gamma correction
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N9/00—Details of colour television systems
    • H04N9/77—Circuits for processing the brightness signal and the chrominance signal relative to each other, e.g. adjusting the phase of the brightness signal relative to the colour signal, correcting differential gain or differential phase
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10024—Color image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20172—Image enhancement details
    • G06T2207/20208—High dynamic range [HDR] image processing
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G2320/00—Control of display operating conditions
    • G09G2320/02—Improving the quality of display appearance
    • G09G2320/0271—Adjustment of the gradation levels within the range of the gradation scale, e.g. by redistribution or clipping
    • G09G2320/0276—Adjustment of the gradation levels within the range of the gradation scale, e.g. by redistribution or clipping for the purpose of adaptation to the characteristics of a display device, i.e. gamma correction
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G2340/00—Aspects of display data processing
    • G09G2340/04—Changes in size, position or resolution of an image
    • G09G2340/0407—Resolution change, inclusive of the use of different resolutions for different screen areas
    • G09G2340/0428—Gradation resolution change
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G2340/00—Aspects of display data processing
    • G09G2340/06—Colour space transformation
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G2360/00—Aspects of the architecture of display systems
    • G09G2360/16—Calculation or use of calculated indices related to luminance levels in display data
    • G—PHYSICS
    • G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
    • G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
    • G09G2370/00—Aspects of data communication
    • G09G2370/04—Exchange of auxiliary data, i.e. other than image data, between monitor and graphics controller

Definitions

  • the present application relates to the field of multimedia communications, and in particular, to a method and a device for video signal processing.
  • High Dynamic Range (HDR) video technology expands the brightness range of the image that can be displayed, so it can record larger brightness range information and show more details of bright and dark parts.
  • HDR is a popular technology that has appeared in the video industry in recent years, and it is also the direction of the future development of the video industry.
  • the real picture seen by the human eye in the real world has a large dynamic range.
  • the traditional Standard Dynamic Range (SDR) display device has low brightness and small dynamic range.
  • Traditional SDR video technology continuously compresses the dynamic range of the captured picture during camera capture, production editing, and encoding.
  • HDR video brightness is usually much larger than the brightness range that SDR display equipment or a large number of other existing HDR display equipment can display; therefore, when displaying HDR video signals on existing display equipment, The device's ability to process the brightness of the HDR video signal to match the range of brightness that the display device can display, suitable for display on existing devices. When the video signal is processed for brightness, unreasonable brightness processing causes the display effect of the HDR video to be ineffective. good.
  • the embodiments of the present application provide a method and a device for processing video signals, which improve the display effect of video signals on a display device.
  • a first aspect of the present application provides a video signal processing method, which includes: obtaining a first linear luminance signal, the first linear luminance signal being obtained based on a first linear red-green-blue RGB signal corresponding to a video signal to be processed; The first linear luminance signal is converted into a first non-linear luminance signal; segmented luminance mapping is performed on the first non-linear luminance signal to obtain a second non-linear luminance signal; and the second non-linear luminance signal is converted into a second non-linear luminance signal.
  • the embodiment of the present application converts the luminance signal to a non-linear space and performs segmented luminance mapping, which can reasonably map the display luminance range of the video signal to the luminance range that can be displayed by the display device, and improve the contrast, brightness and detail performance of the picture, especially In the case of low-brightness display, the display brightness distribution after mapping is reasonable, and the screen display will not be dark; because the brightness mapping is performed in a non-linear space, the errors introduced by the brightness mapping are evenly distributed, and the final display effect of the video signal is The impact is small. In addition, because the brightness range of the video signal to be processed is large, the contribution of video signals in different brightness regions to the video display effect is also different. Different brightness is used based on the characteristics of the brightness interval of the video signal to be processed. The mapping relationship performs segmentation mapping on the brightness value of the video signal to be processed, which improves the flexibility and rationality of the brightness mapping.
  • the video signal to be processed is a perceptually quantized PQ signal
  • obtaining the first linear luminance signal includes: performing color space conversion on the PQ signal to obtain a first non-linear RGB signal; according to the PQ
  • the electro-optical transfer function converts the first non-linear RGB signal into the first linear RGB signal; and performs calculations based on the primary color signals of the first linear RGB signal to obtain the first linear luminance signal.
  • the video signal to be processed is a mixed-log-gamma HLG signal
  • obtaining the first linear luminance signal includes: performing color space conversion on the HLG signal to obtain a second nonlinear RGB signal. ; According to the HLG photoelectric transfer inverse function, convert the second non-linear RGB signal into a second linear RGB signal; perform calculations based on the primary color signals of the second linear RGB signal to obtain a third linear luminance signal; The linear luminance signal is converted into a luminance signal type to obtain the first linear luminance signal.
  • the brightness signal obtained based on the HLG signal is a scene light brightness signal.
  • the scene light brightness signal needs to be converted into a display light brightness signal, and after being converted into a display light brightness signal, it is not directly displayed, but
  • the brightness signal is converted into a non-linear space for segmented brightness mapping, as much as possible to preserve brightness details, improve the rationality of the brightness mapping, and improve the display effect of the HLG signal.
  • the first linear brightness signal is a linear display light brightness signal
  • the third brightness signal is a linear scene light brightness signal
  • the method further includes: performing color space conversion on the RGB display signal to obtain a target display signal, wherein the target display signal is The color format is the same as that of the display device.
  • the method further includes: superimposing a black level level on each primary color value of the RGB display signal to increase BlackLevelLift to obtain processed RGB For a display signal, the BlackLevelLift is the minimum value of the display brightness of the display device.
  • the color space conversion of the RGB display signal includes color space conversion of the processed RGB display signal.
  • the embodiment of the present application considers the influence of the black level of the display device on the brightness mapping curve, and retains the brightness details of the low brightness portion.
  • performing segmented brightness mapping on the first non-linear luminance signal to obtain a second non-linear luminance signal includes: determining a first threshold and a second threshold, the first threshold The threshold value is less than the second threshold value; when the brightness value of the first nonlinear brightness signal is less than or equal to the first threshold value, the brightness value of the second nonlinear brightness signal is equal to the brightness value of the first nonlinear brightness signal; When the brightness value of the first non-linear brightness signal is greater than the first threshold value and less than or equal to the second threshold value, the brightness value of the second non-linear brightness signal is based on the brightness value of the first non-linear brightness signal since A fitting curve of the variable is obtained; when the brightness value of the first non-linear brightness signal is greater than the second threshold, the brightness value of the second non-linear brightness signal is equal to the maximum non-linear display brightness value corresponding to the display device.
  • the video signal to be processed is divided into three segments according to two luminance value thresholds, and a portion smaller than the first threshold is taken as the first segment of the video signal.
  • the luminance after the luminance mapping is equal to the luminance before the luminance mapping, that is, for low luminance Part of the video signal is not compressed, which can retain the image details of the low-brightness part most completely.
  • the high-brightness part is divided into two sections. Among them, the part larger than the first threshold value and smaller than the second threshold value is compressed based on the fitting curve and retained as much as possible. Details of the brightness of this part; parts larger than the second threshold are compressed to the second brightness threshold; segmented mapping of the brightness takes full account of the characteristics of each brightness, retains brightness details as much as possible, and improves the rationality of the brightness mapping.
  • the fitting curve is obtained by performing Hermite interpolation on the first threshold value and the second threshold value.
  • performing segmented brightness mapping on the first non-linear luminance signal to obtain a second non-linear luminance signal includes: using the following segmentation function on the first non-linear luminance signal The signal performs this brightness mapping:
  • e is the first non-linear brightness signal
  • f tm (e) is the second non-linear brightness signal
  • KP1 is the first threshold value
  • KP2 is the second threshold value
  • maxDL is the maximum non-linear display brightness value of the display device
  • MaxSL is the maximum non-linear source luminance value
  • x 0 KP1
  • x 1 maxSL
  • y 0 KP1
  • y 1 maxDL
  • the determining the first threshold and the second threshold includes: determining the first threshold according to a relationship between a display brightness range of the first non-linear brightness signal and a display brightness range of the display device. A threshold value; the maximum brightness value of the first non-linear brightness signal is used as the second threshold value.
  • the selection of the brightness threshold is related to the difference between the source brightness and the display brightness of the display device.
  • the first threshold is equal to the second
  • the threshold is equal to the non-linear maximum brightness value of the source signal.
  • the step of performing segmented brightness mapping on the first nonlinear brightness signal to obtain a second nonlinear brightness signal includes: based on the preset first nonlinear brightness signal. And a mapping relationship with the brightness value of the second non-linear brightness signal to determine the brightness value of the second non-linear brightness signal corresponding to the brightness value of the first non-linear brightness signal.
  • the converting the first linear luminance signal into a first non-linear luminance signal includes: converting the first linear luminance signal into the first non-linear luminance signal according to an inverse PQ electro-optical transfer function.
  • a non-linear luminance signal; correspondingly, converting the second non-linear luminance signal into a second linear luminance signal includes: converting the second non-linear luminance signal into the second linear luminance signal according to a PQ electro-optical transfer function .
  • a second aspect of the present application provides a video signal processing apparatus, which is characterized in that the apparatus includes: a brightness obtaining unit, configured to obtain a first linear brightness signal based on a first linear brightness signal corresponding to a video signal to be processed; A linear red-green-blue RGB signal is obtained; a first conversion unit is configured to convert the first linear brightness signal into a first non-linear brightness signal; a brightness mapping unit is used to perform segmented brightness on the first non-linear brightness signal Mapping to obtain a second non-linear luminance signal; a second conversion unit for converting the second non-linear luminance signal into a second linear luminance signal; a gain calculation unit for calculating the second linear luminance signal and the first linear luminance signal A brightness gain of a linear brightness signal; a display signal acquisition unit is configured to obtain an RGB display signal corresponding to the video signal to be processed based on a product of the brightness gain and the first linear RGB signal.
  • the video signal to be processed is a perceptually quantized PQ signal
  • the brightness obtaining unit is specifically configured to: perform color space conversion on the PQ signal to obtain a first non-linear RGB signal; according to the PQ electro-optic A transfer function converts the first non-linear RGB signal into the first linear RGB signal; and performs calculations based on the primary color signals of the first linear RGB signal to obtain the first linear luminance signal.
  • the video signal to be processed is a mixed logarithmic gamma HLG signal
  • the brightness obtaining unit is specifically configured to perform color space conversion on the HLG signal to obtain a second nonlinear RGB signal; Convert the second non-linear RGB signal into a second linear RGB signal according to the inverse HLG photoelectric transfer function; perform calculations based on the primary color signals of the second linear RGB signal to obtain a third linear luminance signal; The brightness signal is converted into a brightness signal type to obtain the first linear brightness signal.
  • the device further includes: a color space conversion unit, configured to perform color space conversion on the RGB display signal to obtain a target display signal, wherein the color format and display device of the target display signal The corresponding color format is the same.
  • the device further includes: a compensation unit, configured to: superimpose a black level level on each of the primary color values of the RGB display signal and raise BlackLevelLift to obtain a processed RGB display signal, where the BlackLevelLift is the The minimum value of the display brightness of the display device; correspondingly, the color space conversion unit is specifically configured to perform color space conversion on the processed RGB display signal.
  • a compensation unit configured to: superimpose a black level level on each of the primary color values of the RGB display signal and raise BlackLevelLift to obtain a processed RGB display signal, where the BlackLevelLift is the The minimum value of the display brightness of the display device; correspondingly, the color space conversion unit is specifically configured to perform color space conversion on the processed RGB display signal.
  • the brightness mapping unit is specifically configured to determine a first threshold value and a second threshold value, where the first threshold value is less than the second threshold value; and when the brightness value of the first non-linear brightness signal is less than or When equal to the first threshold, the luminance value of the second non-linear luminance signal is equal to the luminance value of the first non-linear luminance signal; when the luminance value of the first non-linear luminance signal is greater than the first threshold value, and is less than or equal to When the second threshold value, the brightness value of the second non-linear brightness signal is obtained based on a fitting curve using the brightness value of the first non-linear brightness signal as an independent variable; when the brightness value of the first non-linear brightness signal is greater than the At the second threshold, the brightness value of the second non-linear brightness signal is equal to the maximum non-linear display brightness value corresponding to the display device.
  • the fitting curve is obtained by performing Hermite interpolation on the first threshold value and the second threshold value.
  • the brightness mapping unit is specifically configured to perform the brightness mapping on the first nonlinear brightness signal by using the following piecewise function:
  • e is the first non-linear brightness signal
  • f tm (e) is the second non-linear brightness signal
  • KP1 is the first threshold value
  • KP2 is the second threshold value
  • maxDL is the maximum non-linear display brightness value of the display device
  • MaxSL is the maximum non-linear source luminance value
  • x 0 KP1
  • x 1 maxSL
  • y 0 KP1
  • y 1 maxDL
  • determining the first threshold and the second threshold includes: determining the first threshold according to a relationship between a display brightness range of the first non-linear brightness signal and a display brightness range of the display device; and The maximum brightness value of the first non-linear brightness signal is used as the second threshold.
  • the brightness mapping unit is specifically configured to determine a relationship with the first non-linear brightness signal based on a preset mapping relationship between the brightness value of the first non-linear brightness signal and the second non-linear brightness signal.
  • the brightness value of the second non-linear brightness signal corresponding to the brightness value of the linear brightness signal.
  • the first conversion unit is specifically configured to convert the first linear luminance signal into the first non-linear luminance signal according to an inverse PQ electro-optical transfer function; correspondingly, the second conversion The unit is specifically configured to convert the second non-linear luminance signal into the second linear luminance signal according to a PQ electro-optical transfer function.
  • a third aspect of the present application provides a video signal processing device, which includes: a processor and a memory; the processor calls software instructions in the memory to perform the following steps: obtaining a first linear luminance signal, the first The linear luminance signal is obtained based on the first linear red-green-blue RGB signal corresponding to the video signal to be processed; converting the first linear luminance signal into a first non-linear luminance signal; and performing segmented luminance mapping on the first non-linear luminance signal, To obtain a second non-linear brightness signal; converting the second non-linear brightness signal into a second linear brightness signal; calculating a brightness gain of the second linear brightness signal and the first linear brightness signal; based on the brightness gain and the first A product of a linear RGB signal to obtain an RGB display signal corresponding to the video signal to be processed.
  • the video signal to be processed is a perceptually quantized PQ signal
  • the processor is specifically configured to: perform color space conversion on the PQ signal to obtain a first nonlinear RGB signal; and perform electro-optical transfer according to the PQ Function to convert the first non-linear RGB signal into the first linear RGB signal; perform calculations based on the primary color signals of the first linear RGB signal to obtain the first linear luminance signal.
  • the video signal to be processed is a mixed log-gamma HLG signal
  • the processor is specifically configured to: perform color space conversion on the HLG signal to obtain a second nonlinear RGB signal; according to the HLG
  • the inverse photoelectric transfer function converts the second nonlinear RGB signal into a second linear RGB signal; calculates based on the primary color signals of the second linear RGB signal to obtain a third linear luminance signal; and the third linear luminance signal Perform a luminance signal type conversion to obtain the first linear luminance signal.
  • the processor is further configured to perform color space conversion on the RGB display signal to obtain a target display signal, wherein the color format of the target display signal is the same as the color format corresponding to the display device.
  • the processor is configured to superimpose a black level level on each primary color value of the RGB display signal and increase BlackLevelLift to obtain a processed RGB display signal, where the BlackLevelLift is a display brightness of the display device.
  • Minimum value correspondingly, the processor is specifically configured to perform color space conversion on the processed RGB display signal.
  • the processor is specifically configured to determine a first threshold value and a second threshold value, where the first threshold value is less than the second threshold value; and when the brightness value of the first non-linear brightness signal is less than or equal to the When the first threshold value, the brightness value of the second nonlinear brightness signal is equal to the brightness value of the first nonlinear brightness signal; when the brightness value of the first nonlinear brightness signal is greater than the first threshold value and less than or equal to the first When the threshold value is two, the brightness value of the second nonlinear brightness signal is obtained based on a fitting curve using the brightness value of the first nonlinear brightness signal as an independent variable; when the brightness value of the first nonlinear brightness signal is greater than the second At the threshold, the brightness value of the second nonlinear brightness signal is equal to the maximum nonlinear display brightness value corresponding to the display device.
  • the fitting curve is obtained by performing Hermite interpolation on the first threshold value and the second threshold value.
  • the fitted curve is stored in the memory.
  • the processor is specifically configured to perform the brightness mapping on the first non-linear brightness signal by using the following piecewise function:
  • e is the first non-linear brightness signal
  • f tm (e) is the second non-linear brightness signal
  • KP1 is the first threshold value
  • KP2 is the second threshold value
  • maxDL is the maximum non-linear display brightness value of the display device
  • MaxSL is the maximum non-linear source luminance value
  • x 0 KP1
  • x 1 maxSL
  • y 0 KP1
  • y 1 maxDL
  • the piecewise function is stored in the memory.
  • the processor is specifically configured to: determine the first threshold value according to a relationship between a display brightness range of the first non-linear brightness signal and a display brightness range of the display device; The maximum brightness value of the brightness signal is used as the second threshold.
  • the processor is specifically configured to determine a relationship with the first non-linear brightness based on a preset mapping relationship between the first non-linear brightness signal and the second non-linear brightness signal ’s brightness value.
  • the brightness value of the signal corresponds to the brightness value of the second non-linear brightness signal.
  • the mapping relationship is stored in the memory.
  • the processor is specifically configured to convert the first linear luminance signal into the first non-linear luminance signal according to an inverse PQ electro-optical transfer function; correspondingly, the processor is specifically configured to : Converting the second non-linear luminance signal into the second linear luminance signal according to a PQ electro-optical transfer function.
  • a fourth aspect of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores instructions, and when it runs on a computer or a processor, the computer or processor executes the first aspect as described above. Or the method described in any of its possible embodiments.
  • a fifth aspect of the present application provides a computer program product containing instructions, which when executed on a computer or processor, causes the computer or processor to execute the first aspect or any possible implementation manner described above. The method described.
  • FIG. 1 is a schematic diagram of an exemplary application scenario provided by an embodiment of the present application.
  • FIG. 2 is a schematic structural diagram of an exemplary application scenario according to an embodiment of the present application.
  • FIG. 3 is a schematic structural diagram of another exemplary application scenario according to an embodiment of the present application.
  • FIG. 4 is a schematic diagram of a hardware architecture of an exemplary playback device and display device according to an embodiment of the present application
  • FIG. 5 is a flowchart of an exemplary video signal processing method according to an embodiment of the present application.
  • FIG. 6 is a flowchart of an exemplary method for processing brightness of an HDR video signal according to an embodiment of the present application
  • FIG. 7 is a schematic diagram of an exemplary brightness mapping curve provided by an embodiment of the present application.
  • FIG. 8 (a) is a schematic diagram of an exemplary static metadata HDR tone mapping provided by an embodiment of the present application.
  • FIG. 8 (b) is a schematic diagram of an exemplary dynamic metadata HDR tone mapping provided by an embodiment of the present application.
  • FIG. 9 is a flowchart of an exemplary method for obtaining a brightness mapping curve according to an embodiment of the present application.
  • FIG. 10 is a flowchart of another exemplary HDR signal brightness processing method according to an embodiment of the present application.
  • FIG. 11 is an exemplary PQ EOTF curve (left) and an exemplary PQ EOTF -1 curve (right) provided by an embodiment of the present application;
  • FIG. 12 is an exemplary HLG OETF curve (left) and an exemplary HLG OETF -1 curve (right) provided by an embodiment of the present application;
  • FIG. 13 is an exemplary apparatus for processing a video signal according to an embodiment of the present application.
  • FIG. 14 is an exemplary HDR terminal technical solution processing flowchart according to an embodiment of the present application.
  • 15 is a schematic diagram of an exemplary color gamut conversion process according to an embodiment of the present application.
  • 16 is a schematic diagram of an exemplary signal conversion process according to an embodiment of the present application.
  • 17 is a schematic diagram of an exemplary test network provided by an embodiment of the present application.
  • FIG. 18 is a schematic diagram of another exemplary test network provided by an embodiment of the present application.
  • FIG. 20 is a schematic diagram of an exemplary 1000 cd / m2 HLG curve HDR video end-to-end system according to an embodiment of the present application.
  • FIG. 21 is an exemplary non-1000cd / m2 HLG curve HDR video end-to-end system provided by an embodiment of the present application.
  • At least one (a), a, b, or c can represent: a, b, c, "a and b", “a and c", “b and c", or "a and b and c" ", Where a, b, and c can be single or multiple.
  • Color value a value corresponding to a specific image color component (such as R, G, B or Y).
  • Digital code value A digital expression value of an image signal.
  • a digital code value is used to represent a non-linear primary color value.
  • Linear color value (linear color value): The linear color value is proportional to the light intensity. In an optional case, its value should be normalized to [0,1], referred to as E.
  • Non-linear primary color value is the normalized digital expression value of the image information, which is proportional to the digital encoding value. In an optional case, its value should be normalized to [0,1], referred to as E ′.
  • Electro-optical transfer function A conversion relationship from a non-linear primary color value to a linear primary color value.
  • Optical-electric transfer function A conversion relationship from a linear primary color value to a non-linear primary color value.
  • Metadata Data describing the video source information carried in the video signal.
  • Dynamic metadata Metadata associated with each frame of image. This metadata changes from picture to picture.
  • Static metadata Metadata associated with an image sequence, which remains unchanged within the image sequence.
  • Luma signal (luma) represents a combination of non-linear primary color signals, the symbol is Y '.
  • Brightness mapping The brightness of the source image is mapped to the brightness of the target system.
  • Chroma volume The volume formed by the chromaticity and brightness of a display in chromaticity space.
  • Display adaptation processing video signals to adapt to the display characteristics of the target display.
  • Source image The image input during the HDR pre-processing stage.
  • Mastering Monitor (Mastering Display): The reference display used in the editing and production of video signals to determine the effect of video editing and production;
  • HDR video signal with content as scene light in HDR video technology refers to the scene light captured by the camera / camera sensor, which is generally a relative value; after the linear scene light signal is HLG encoded The HLG signal is obtained.
  • the HLG signal is a scene light signal, and the HLG signal is nonlinear.
  • the scene light signal generally needs to be converted to a display light signal through OOTF and displayed on a display device.
  • Linear display light signal that uses content as display light in HDR video technology refers to the display light emitted by the display device, generally an absolute value, in units of nits; the linear display light signal is PQ coded After that, a PQ signal is obtained.
  • the PQ signal is a display light signal and the PQ signal is a non-linear signal.
  • the general standard of the display light signal is displayed on the display device according to its absolute brightness.
  • OOTF Light-to-light conversion curve
  • Dynamic Range The ratio of the maximum brightness to the minimum brightness in a video signal
  • Luma-Chroma-Chroma Bright color separates the three components of the video signal
  • PQ Perceptual Quantizer
  • PQEOTF curve converts the PQ-coded electrical signal into a linear optical signal, in units of nits; the conversion formula is:
  • E ′ is the input electric signal, and the value range is [0,1]; the fixed parameter values are as follows:
  • the PQEOTF curve is shown in the left figure of Figure 11: the input is an electrical signal in the range [0,1], and the output is a linear optical signal at [0,10000] nits;
  • PQ EOTF -1 curve the inverse curve of PQ EOTF; the physical meaning is to convert the linear optical signal of [0,10000] nits into a PQ-coded electrical signal; the conversion formula is:
  • the PQ EOTF -1 curve is shown in the right figure of Figure 11.
  • the input is a linear optical signal with [0,10000] nits, and the output is an electrical signal with a range of [0,1];
  • Color Gamut A color space contains the range of colors.
  • the relevant color gamut standards are BT.709, BT.2020.
  • Hybrid Log Gamma An HDR standard. Video signals collected by cameras, cameras, image sensors, or other types of image acquisition equipment are video signals in the HLG encoding format.
  • HLG OETF curve A curve that transforms a linear scene light signal into a non-linear electrical signal by HLG encoding.
  • the conversion formula is as follows:
  • E is the input linear scene light signal, the range [0,1];
  • E ′ is the output non-linear electrical signal, the range [0,1];
  • HLG OETF -1 curve The inverse curve of HLG OETF converts HLG-coded non-linear electrical signals into linear scene light signals.
  • the conversion formula is as follows:
  • E ′ is the input non-linear electrical signal with a range of [0,1]
  • E is the output linear scene light signal with a range of [0 ,1].
  • Linear space In this application, linear space refers to the space where the linear optical signal is located
  • Non-linear space refers to the space where the linear optical signal is transformed by using a non-linear curve; the non-linear curves commonly used in HDR include PQ, EOTF-1, HLG, OETF, etc.
  • the linear curve has a gamma curve; it is generally considered that the linear optical signal is visually linear with respect to the human eye after being encoded by the above-mentioned non-linear curve. It should be understood that the non-linear space can be considered as a visual linear space.
  • Gamma correction is a method for non-linear tone editing of an image. It can detect the dark and light parts of the image signal and increase the ratio of the two, thereby improving the contrast of the image. .
  • the non-linear conversion of the color values output by the device is because the human visual system is not linear, and humans perceive visual stimuli through comparison.
  • the outside world strengthens the stimulus by a certain proportion. For people, this stimulus is evenly increased. Therefore, for human perception, the physical quantity added in a series of equal proportions is uniform.
  • the value of gamma can be determined according to the photoelectric conversion curve of the color space.
  • Color space can be the eyes' different perception of light at different frequencies, and it can also represent objectively existing light at different frequencies.
  • Color space is a color range defined by the coordinate system that people establish to represent color.
  • the color gamut together with the color model, defines a color space.
  • the color model is an abstract mathematical model that uses a set of color components to represent colors.
  • the color model may include, for example, a three-primary-color light mode (red, green, blue, RGB), and a printing four-color mode (cyan, magenta, yellow, or key plate) (CMYK).
  • Color gamut is the sum of colors that a system can produce.
  • AdobeRGB and sRGB are two different color spaces based on the RGB model.
  • Each device such as a monitor or printer, has its own color space and can only generate colors within its color gamut.
  • the color of the image on different devices may change because each device converts and displays RGB or CMYK according to its own color space.
  • sRGB color space (standard Red Green Blue color space): is a standard RGB color space developed by HP and Microsoft in 1996 for monitors, printers and the Internet. It provides a standard way to define color, allowing a variety of computer peripherals and applications, such as display, print, and scan, to have a common language for color.
  • the sRGB color space is based on independent color coordinates, which can make colors correspond to the same color coordinate system in the transmission of different devices without being affected by different color coordinates of these devices.
  • the color gamut space of sRGB is relatively small. sRGB defines the colors of the three primary colors of red, green, and blue.
  • the color value of one of the three primary colors takes the maximum value, and the color corresponding to the color value of the other two colors is zero represents the one color.
  • the values of the color values R, G, and B are all 0-255, and when the values of R and G are all zero, and the value of B is 255, Color indicates blue.
  • YCC color space The color space for bright color separation in this application.
  • the three components of YCC represent Luma-Chroma-Chroma respectively.
  • Common YCC space video signals include YUV, YCbCr, ICtCp, etc .;
  • Reserved_bits "Reserved bits" in the bit stream indicate that some syntax units are reserved for future extensions to this section. These bits should be ignored during decoding. The "reserved bits” should not have more than 21 consecutive '0's starting from any byte-aligned position.
  • Marker bit (marker_bit): It means that the value of this bit should be '1'.
  • the HDR video can display a large brightness.
  • the maximum brightness of the SDR video signal is 100nits, and the maximum brightness of the HDR video signal is more than 1000nits.
  • a large number of existing display devices can display a brightness range that is less than that of HDR video. Brightness; therefore, when displaying HDR video signals, the brightness of the HDR video signal needs to be processed according to the display capability of the display device, so that the HDR video signal matches the range of brightness that the display device can display, suitable for display on current devices; ITU.BT.2100
  • the defined PQ signal and HLG signal are two internationally recognized HDR signal sources, and have been included in the standard by many countries and regions.
  • FIG. 1 it is a schematic diagram of an exemplary application scenario provided by an embodiment of the present application.
  • the playback device 102 finishes receiving and decoding the video stream 101, and the playback device 102 sends the decoded video or audio data to the display device 103 for display through a high definition multimedia interface (High Definition Multimedia Interface). Or play so that users can enjoy video or audio content.
  • a high definition multimedia interface High Definition Multimedia Interface
  • the video stream 101 may come from a website streaming media, a remote network device, the Internet, an optical fiber network, and the like.
  • the video stream may be dynamic metadata HDR video data or static metadata HDR video data.
  • the video stream 101 may be a data stream in the Transport Stream (TS) format.
  • the TS may include a video stream, an audio stream, a subtitle data packet, and the like.
  • the video stream may also use other similar format data streams, such as streaming media. You can use the (Matroska Video File, MKV) format to encapsulate audio data, video data, and subtitle data at the same time.
  • MKV Meatroska Video File
  • the transmission format of the audio and video streams in this application is not Make a limitation; for example, the video stream may include: HDR video data and metadata for describing the HDR video. In this case, both metadata and HDR video data are compressed in the video stream.
  • the TS may include a video stream, an audio stream, a subtitle data packet, and metadata used to describe the HDR video. In this case, the metadata used to describe the HDR video data is placed in the TS. Without compression in the video stream.
  • the metadata contains a description of the video image data.
  • the static metadata describes the production environment of the entire video, which may include information about the monitor used for color adjustment and correction of video production, peak brightness, black level, RGB three-color coordinates, and white point coordinates.
  • the dynamic metadata generally includes a description of each frame of the video image, for example, it may include the highest brightness, the lowest brightness, and the average brightness of the image; optionally, the dynamic metadata may also include a certain frame of image
  • the reference mapping curve of the display screen it should be understood that the reference mapping curve included in the dynamic metadata varies with the change of the video image.
  • the playback device 102 may be a set-top box (STB), a multimedia player, etc.
  • the STB mainly includes more security functions, such as card charging, video encryption and decryption, etc.
  • Some high-quality videos have a digital rights protection mechanism, which needs to be decrypted by the STB side before they can be viewed on the TV side.
  • the video data received by the STB side is usually encoded video data.
  • the STB also has a decoding function. After the data is decoded, it is sent to the TV side for display.
  • FIG. 3 a schematic diagram of an application scenario in which a playback device completes processing video data and then sends it to a display device according to an embodiment of the present application.
  • the STB parses the received TS to obtain video data. , Audio data, metadata metadata, etc.
  • the video data received by the STB may be an HDR HLG video signal, an HDR PQ video signal, or an SDR video signal.
  • the STB decodes the video data, displays brightness processing, color saturation processing, color gamut processing, etc., so that the HDR video signal adapts to the brightness range of the display screen, and then passes the processed video data through wired or wireless HDMI, Display Port transmission to the display device 103.
  • the video data processed by the playback device matches the brightness range that the display device can display.
  • the type of processed video data is related to the type of display device and can be HDR.
  • the video data may also be SDR video data. As shown in FIG.
  • the calculation of the brightness mapping curve is implemented by the main processor and is generally completed by software; the brightness mapping curve calculated in advance is stored in the memory, and the brightness processing unit in the video processor calls the brightness mapping curve in the memory.
  • the video processor may be dedicated integrated hardware or dedicated circuits or multiple software modules running on a dedicated chip; it should be understood that FIG. 3 is only an example for a playback device.
  • the architecture is not limited. In actual use, the implementation of each functional module can be adjusted according to the actual application scenario. It can be implemented by either the video processor or the main processor software, or other dedicated chips such as DSP, FPGA and other hardware are combined with software.
  • a video stream or the like containing video data may be directly transmitted to a display device 103, such as a TV, etc., and then the video data is subjected to video decoding, brightness mapping processing, color saturation processing, and color gamut within the display device.
  • a display device 103 such as a TV, etc.
  • the video data is subjected to video decoding, brightness mapping processing, color saturation processing, and color gamut within the display device.
  • FIG. 2 is a schematic diagram of an application scenario architecture provided by a display device after processing video data by a display device according to an embodiment of the present application.
  • the calculation of the brightness mapping curve is realized by the main processor, which is generally completed by software instructions; the brightness mapping curve calculated in advance is stored in the memory, and the brightness processing unit in the video processor calls the memory
  • the brightness mapping curve completes the processing of the video picture; it should be understood that in actual use, the implementation of each functional module can be adjusted according to the actual application scenario, which can be implemented by either the video processor or the main processor software. It can also be implemented by using other special-purpose chips such as DSP, FPGA and other hardware combined with software.
  • the display device 103 may be a TV, a computer monitor, or any fixed terminal or mobile terminal having a display function.
  • TV displays the video data sent by STB and displays it on the screen.
  • the display processing may be that the TV adjusts the brightness of the video according to a preset brightness mapping relationship, so that the video content is compatible with the display capability of the TV.
  • the TV uses different sets of brightness mapping relationships contained in the dynamic metadata to different videos.
  • the picture is processed for brightness mapping tone mapping so that various pictures can be displayed and presented in the best way.
  • the TV may also be an SDR TV or an HDR PQ TV.
  • the following describes a playback device and a display device in the embodiments of the present application from the perspective of processor hardware.
  • FIG. 4 it is a schematic diagram of a hardware architecture of a playback device and a display device according to an embodiment of the present application.
  • the playback device 102 includes at least one central processing unit CPU, at least one memory, GPU, decoder, dedicated video / graphics processor, input interface, HDMI transmitter, and the like.
  • the playback device may further include a microprocessor, a microcontroller (Microcontroller Unit, MCU), and the like.
  • MCU microcontroller Unit
  • the foregoing parts of the playback device are coupled through a connector, and the connector may include various interfaces, transmission lines, or buses, which is not limited in this embodiment.
  • the connector may include various interfaces, transmission lines, or buses, which are not limited in this embodiment.
  • the above parts are integrated on the same chip and together form the core chip of the playback device; in another optional case, the CPU, GPU, decoder, input interface, and HDMI transmitter Integrated on a chip, the internal parts of the chip access the external memory through the bus.
  • the dedicated video / graphics processor can be integrated on the same chip as the CPU, or it can exist as a separate processor chip.
  • the chip involved in the embodiment of the present application is a system manufactured on the same semiconductor substrate by an integrated circuit process, also called a semiconductor chip, which can be manufactured on the substrate (usually, for example, silicon or the like) by using an integrated circuit process. Semiconductor material), the outer layer of which is usually encapsulated by a semiconductor packaging material.
  • the integrated circuit may include various types of functional devices, and each type of functional devices includes logic gate circuits, Metal-Oxide-Semiconductor (MOS) transistors, bipolar transistors, or transistors such as diodes, and may also include capacitors and resistors. Or other components such as inductors.
  • MOS Metal-Oxide-Semiconductor
  • bipolar transistors bipolar transistors
  • transistors such as diodes
  • capacitors and resistors or other components such as inductors.
  • Each functional device can work independently or under the action of necessary driver software, and can achieve various functions such as communication, calculation, or storage.
  • the CPU may be used to implement part or all of the operations in the embodiments of the present application, for example, it may implement tone mapping of an image, demultiplexing and decoding of video data, calculation of a brightness-mapping curve, and a color saturation curve. Calculation, etc .; optionally, the CPU may be a single-CPU processor or a multi-CPU processor; optionally, the CPU may be a processor group composed of multiple processors, and multiple The processors are coupled to each other through one or more buses. In an optional case, the processing of images and videos is partly performed by the GPU, partly by a dedicated video / graphics processor, and possibly by software code running on a general-purpose CPU or GPU.
  • Memory which can be used to store computer program instructions, including various operating system (Operation System, OS), various user application programs, and various computer program codes including program code for executing the scheme of the present application.
  • OS Operating System
  • various user application programs and various computer program codes including program code for executing the scheme of the present application.
  • it can be used to store brightness Mapping curve or brightness mapping lookup table LUT;
  • the memory can also be used to store video data, audio data, subtitle data, etc .;
  • the CPU can be used to execute computer program code stored in the memory to implement the method in the embodiment of the present application and be executed
  • the various types of program code can be considered as the driver of the CPU.
  • the memory 302 may be a non-power-loss volatile memory, such as an Embedded Multimedia Card (EMMC), a Universal Flash Memory (UFS), or a Read-Only Memory (Read-Only Memory, ROM), or other types of static storage devices that can store static information and instructions, can also be volatile memory (volatile memory), such as Random Access Memory (Random Access Memory, RAM) or can store information and Other types of dynamic storage devices for instructions can also be Electrically Erasable Programmable Read-Only Memory (EEPROM), Compact Disc Read-Only Memory (CD-ROM), or other optical disks Storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or can be used to carry or store program code in the form of instructions or data structures and can Any other computer-readable storage medium accessed by the computer, but is not limited thereto.
  • EMMC Embedded Multimedia Card
  • UFS Universal Flash Memory
  • the input interface is used to receive the transport stream.
  • the input interface of the playback device can be a network interface, such as a WIFI or Ethernet interface.
  • the input interface can also be a terminal of a broadcast television such as a tuner.
  • the input interface can also be a universal serial bus. (Universal Serial Bus, USB) interface.
  • the CPU executes related code to demultiplex the transport stream to obtain video data and subtitle data, etc.
  • the decoder decodes the video data stream to obtain video data and metadata
  • the video / graphics processor completes For video data brightness mapping processing, color saturation processing, color space conversion, color preprocessing, scene light and display light signal conversion, linear space and non-linear space conversion, etc., optionally, it can also complete the brightness mapping curve Calculation of saturation mapping curve.
  • the HDMI transmitter encapsulates the decoded video data, metadata and subtitle data respectively, and transmits the encapsulated data packets / information frames to the display device 103 through the HDMI data channel.
  • the display device 103 includes an HDMI receiver, at least one central processing unit CPU, at least one memory, a GPU, a decoder, a dedicated video / graphics processor, and a ByOne interface.
  • the display device also includes a display screen (not shown in the figure).
  • the VByOne interface is coupled to the display screen.
  • the VByOne interface is a digital interface standard developed for image transmission.
  • the display device 103 shown in FIG. 4 may be an integrated display chip, and the received video data is processed on the display chip and sent to a display screen for display.
  • the HDMI transmitter separates video data frames, metadata information frames, subtitle information frames, and other information frames or data packets and transmits them to the display device. It should be understood that there are multiple channels inside HDMI, some of which are used to transmit data information, and some are used to transmit control information such as clock, check signal, power signal, and ground signal.
  • the data channel is time-multiplexed, and various data packets cannot be transmitted simultaneously.
  • the amount of data transmitted in a unit time of a channel is limited by the operating frequency.
  • the maximum amount of data that can be transmitted in a unit time of a channel is the bandwidth of the HDMI channel.
  • the bandwidth of HDMI 2.1 is 18Gbps (bit second).
  • the transmission interface transmits the HDR video data frame, metadata information frame, and subtitle information frame in a time-sharing manner.
  • the transmission interface corresponds to multiple transmission frequency bands, and the transmission interface divides and transmits video data frames, metadata frames, and subtitle information frames; optionally, the transmission interface corresponds to multiple transmission channels, and the transmission interface transmits video data frames in different channels. Metadata frame and subtitle information frame.
  • Tone mapping of video data can be done by the GPU or by a dedicated video / graphics processor.
  • Luminance mapping can be done by a dedicated video / graphics processor, or by software code running on a CPU or GPU.
  • the video / image processor transmits the video data after display brightness processing to the display screen for display through the VByOne interface.
  • the display screen may be a liquid crystal display (Liquid Crystal Display, LCD), a light emitting diode (Light Emitting Diode, LED) display, an organic light emitting diode (Organic Light-Emitting Diode, OLED) display, or a cathode ray tube ( Cathode Ray Tube (CRT) display and so on.
  • LCD Liquid Crystal Display
  • LED Light Emitting Diode
  • OLED Organic Light-Emitting Diode
  • CTR Cathode Ray Tube
  • FIG. 5 it is a flowchart of a video signal processing method according to an embodiment of the present application. It should be understood that for the convenience of description, FIG. 5 describes the method in the form of steps. Although the method sequence is shown in the method flowchart 5, in some cases, the description may be performed in a different order than here. A step of.
  • the video signal processing method includes:
  • the first linear luminance signal is obtained based on a first linear RGB signal corresponding to a video signal to be processed.
  • the first linear RGB signal is a linear display light signal
  • the first linear luminance signal is a linear display light luminance signal.
  • the luminance signal is a luminance component of a video signal to be processed.
  • the first linear luminance signal is calculated based on the three primary color signals R, G, and B of the first linear RGB signal.
  • the video signal to be processed may be a PQ signal
  • the PQ signal may be a PQ signal in a YUV space.
  • the PQ signal needs to be converted from the YUV space to the RGB space to obtain a first nonlinear RGB signal.
  • the first non-linear RGB signal is converted into the first linear RGB signal, and then calculation is performed based on each primary color signal of the first linear RGB signal to obtain a first linear luminance signal.
  • the first linear luminance signal is a display luminance signal.
  • the video signal to be processed may be an HLG signal
  • the HLG signal may be an HLG signal in a YUV space.
  • the HLG signal needs to be converted from the YUV space to the RGB space to obtain a second nonlinear RGB signal
  • the second non-linear RGB signal is converted into a second linear RGB signal
  • the second linear RGB signal is a linear scene light signal, and then based on each primary color signal of the second linear RGB signal Perform calculation to obtain the third linear luminance signal.
  • the third linear luminance signal needs to be converted into a luminance signal type to obtain the first linear luminance signal.
  • Linear brightness signal is a display light brightness signal.
  • the embodiment of the present application performs brightness mapping in a non-linear space.
  • the input of the brightness mapping is a linear brightness signal, so the linear brightness signal needs to be converted into a non-linear brightness signal.
  • the first linear luminance signal may be converted into the first non-linear luminance signal according to an inverse PQ electro-optical transfer function.
  • Other conversion functions or conversion curves may also be used to complete the conversion between the first linear luminance signal and the first non-linear luminance signal.
  • the first non-linear brightness signal is a non-linear brightness signal before brightness mapping.
  • segmented brightness mapping may be performed on the first non-linear brightness signal based on a preset brightness mapping lookup table.
  • the brightness mapping lookup table may be stored in a memory, and the brightness mapping lookup table includes several discrete sets of coordinate points.
  • the mapping data in the brightness mapping lookup table may be obtained through prior experiments;
  • the first non-linear luminance signal may be segmented in luminance mapping based on a piecewise function.
  • the first non-linear luminance signal may be divided into three segments according to the luminance value, and the luminance of each segment is The mapping functions are all different. For example, the first threshold value and the second threshold value are determined, and the first threshold value is smaller than the second threshold value. When the brightness value of the first nonlinear brightness signal is less than or equal to the first threshold value, the second nonlinear brightness signal is determined.
  • the brightness value of is equal to the brightness value of the first non-linear brightness signal; when the brightness value of the first non-linear brightness signal is greater than the first threshold value and less than or equal to the second threshold value, the brightness value of the second non-linear brightness signal is based on the
  • the brightness value of a non-linear brightness signal is obtained by a fitting curve of an independent variable; when the brightness value of the first non-linear brightness signal is greater than a second threshold, the brightness value of the non-linear brightness signal is equal to the maximum non-linearity corresponding to the display device Displays the brightness value.
  • the fitted curve is obtained by performing Hermite interpolation on the first threshold and the second threshold.
  • the piecewise brightness mapping curve can be expressed as the following piecewise function:
  • the piecewise brightness mapping curve can be expressed as the following piecewise function:
  • the piecewise brightness mapping curve can be expressed as the following piecewise function:
  • the piecewise brightness mapping curve can be expressed as the following piecewise function:
  • the piecewise brightness mapping curve can be expressed as the following piecewise function:
  • the piecewise brightness mapping curve can be expressed as the following piecewise function:
  • the piecewise brightness mapping curve can be expressed as the following piecewise function:
  • segmented brightness mapping may be performed on the first non-linear brightness signal based on the brightness mapping curve, and the brightness mapping curve may be a segmented curve. It should be understood that the segmented brightness mapping curve may be considered as an illustration of a segmented function.
  • the discrete data in the lookup table may be coordinate points on a brightness mapping curve.
  • the second non-linear luminance signal may be converted into the second linear luminance signal according to the PQ electro-optical transfer function.
  • Other conversion functions or conversion curves may also be used to complete the conversion of the second non-linear luminance signal and the second linear luminance signal.
  • the second non-linear brightness signal is a non-linear brightness signal after brightness mapping.
  • the brightness gain is a ratio of the second linear brightness signal to the first linear brightness signal.
  • the brightness gain is multiplied with the three primary color components R, G, and B of the first linear RGB signal to obtain an RGB display signal.
  • the RGB display signal can be used for display on a display device.
  • the display device can display a color format different from RGB, and the method further includes:
  • a color space conversion is performed on the RGB display signal to obtain a target display signal.
  • the color format of the target display signal is the same as the color format corresponding to the display device.
  • RGB display signal after obtaining the RGB display signal, superimpose the black level level on each of the primary color components R, G, and B of the RGB display signal to raise the BlackLevelLift, and BlackLevelLift is the minimum value of the display brightness of the display device. . Further, color space conversion is performed on the processed display RGB signal to obtain a target display signal with the same color format as that of the display device.
  • the video signal processing method provided in the embodiment of the present application is described below by taking the input as an HDR PQ signal and the input as an HDR HLG signal as examples.
  • FIG. 6 it is a schematic flowchart of a method for processing brightness of HDR video data according to an embodiment of the present application.
  • the linear scene light signal and the linear display light signal are two types of video signals.
  • the linear scene light signal is a video signal captured by a camera or other image acquisition device or video acquisition device, and the linear display light signal is displayed by the display device.
  • the luminance signal is the component that represents the luminance in the video signal.
  • a linear scene luminance signal is obtained based on the linear scene light signal, and a linear display light luminance signal is obtained based on the linear display light signal.
  • the method includes:
  • the brightness mapping curve may be any brightness mapping curve acting on the selected non-linear space.
  • the non-linear space may be a PQ curve space.
  • Figure 7 shows an example of a brightness mapping curve generated in a non-linear space (PQ EOTF -1 curve).
  • the horizontal axis of the brightness mapping curve is a non-linearly encoded brightness signal before the brightness mapping, and the vertical axis is the brightness mapping.
  • the range of the abscissa and ordinate values are both [0,1], which represents the electric power after the linear light signal with the luminance range of [0,10000] nits is encoded by the PQ curve.
  • the range of the signal is [0,1] (refer to the right figure in Figure 11, the brightness signal in the brightness range of 0-10000nits is encoded by the PQ curve to obtain the electrical signal of 0-1), that is, the abscissa represents the value before the brightness mapping.
  • the brightness range of the brightness signal is [0,10000] nits
  • the ordinate represents the brightness range of the brightness signal after the brightness mapping
  • the curve shown in FIG. 7 represents mapping the brightness signal with the brightness range of [0,10000] nits to the brightness range. [0,300] nits.
  • the curve of the brightness mapping is a piecewise function.
  • the formula of the brightness mapping curve can be as follows:
  • a one-dimensional lookup table LUT is used to represent the brightness mapping curve.
  • yn is the ordinate of the n sampling points, that is, the output of the brightness mapping curve, and represents the brightness signal after the brightness mapping. It should be understood that since the mapping relationship in the lookup table is a set of discrete points, if the input brightness value is not among these discrete points, the input brightness value can be obtained by interpolation based on known brightness values; in an optional In the case, the luminance values of the inputs within a range correspond to the luminance values of the same output, for example, x0-x3 all correspond to y0, x4-x6 all correspond to y1, and so on.
  • These discrete coordinate points can be obtained through prior experimental measurements, such as inputting the brightness value, and measuring the display brightness value corresponding to the input brightness value on the display screen; in an optional case, the input brightness value and the The fitting function of the corresponding relationship of the brightness values displayed on the display screen, and the discrete coordinate points in the lookup table are calculated according to the fitting function.
  • the essence of brightness mapping is to adjust the display brightness range of a video image, and map from one display brightness range to another display brightness range.
  • the brightness range of a video image can be adjusted to a brightness that can be displayed on a display screen.
  • the input signal of the TM_Curve curve is the brightness signal before brightness mapping
  • the output is the brightness signal after brightness mapping.
  • tone mapping can be an implementation of brightness mapping, or tone mapping In some cases, the process can be equivalent to the process of brightness mapping.
  • the static metadata HDR video uses a fixed mapping method to process all video images.
  • the static metadata HDR video has a fixed mapping curve.
  • the input brightness range of the mapping curve is 0-4000 nits.
  • Special nit the output brightness range is 0-500nit, TV for the maximum brightness of 200nit, the maximum brightness of 1500nit and the maximum brightness of 4000nit three images, all use this mapping curve for tone mapping, after processing and display on the display. Because the mapping relationship curve does not match the actual brightness range of the first two frames of the image, most of the image details are lost in the first two frames of the image, the overall image is dark, and the display effect is poor.
  • Dynamic metadata HDR video has multiple sets of mapping curves.
  • mapping curves used are adjusted according to the actual brightness of the image.
  • Figure 8 (b) three different mapping curves are given.
  • the input brightness range for curve 1 is 0-500 nits, and the output brightness range is 0-500nits.
  • the brightness range for input curve 2 is 0-1500 nits, and the output brightness range is 0-500 nit.
  • Mapping curve 3 The input brightness range is 0-4000 nits, and the output brightness range is 0-500nits.
  • TV selects a suitable mapping curve for tone mapping based on the actual brightness range of the image frame, so that images of different brightness can be optimally used. It is displayed on the screen.
  • the mapping curve 2 is selected to perform brightness mapping on an image with a maximum brightness of 1500nit.
  • the processed image retains the image details well.
  • the input brightness of the above mapping curve represents the brightness represented by the video signal source; and the output brightness is the brightness that a display device such as a TV can actually display.
  • the brightness range represented by video signals produced in a professional film and television production environment is generally larger than the brightness range that consumer televisions can display.
  • Tone mapping is a technology that maps and matches the brightness range of an input video signal to the brightness range displayed by a display device.
  • the input video signal in the embodiment of the present application is an HDR PQ signal in the YUV space.
  • PQ coding is performed on the linear display light signal to obtain a PQ signal
  • the PQ signal is a display light signal
  • the PQ signal is a kind of non-linear signal.
  • the input video signal is converted from YUV space to RGB space to obtain the non-linear display light signal R'dG'dB'd in the RGB color space; further, the non-linear display light signal R'dG ' dB'd is converted into a linear display optical signal RdGdBd.
  • the nonlinear display optical signal R'dG'dB'd is converted into a linear display optical signal RdGdBd based on the PQ EOTF curve.
  • the conversion from YUV to RGB is considered to be a color space conversion, and the non-linear RGB to linear RGB signals are not converted in color space.
  • the non-linear RGB signals and the linear RGB signals Both belong to the RGB color space.
  • converting the input video signal from a YUV signal to a non-linear display light signal R'dG'dB'd is a color space conversion to convert the non-linear display light signal R
  • the conversion of 'dG'dB'd to the linear display light signal RdGdBd can also be considered as a color space conversion.
  • the nonlinear RGB display light signal and the linear RGB display light signal belong to different color spaces.
  • Yd is a linear luminance signal
  • the luminance signal is the amount that represents the brightness of the display light signal, that is, the luminance signal can also be considered as the component that represents the brightness of the video signal; among which, the selection of the parameters cr, cg, cb and the linear display optical signal RdGdBd
  • the corresponding calculation parameters will also be different. Therefore, when calculating the brightness, it is necessary to select the linear brightness calculation parameter in the corresponding color gamut according to the color gamut in which the light signal is located.
  • Yd is the display brightness in a linear space.
  • the display brightness Yd is converted to a non-linear space using a PQ EOTF -1 curve to obtain a non-linear display brightness.
  • NL_Yd PQ_EOTF -1 (Yd), which is equivalent to
  • the linear display light luminance signal Yd with a luminance range of 0-10,000 nits is encoded by a PQ curve to obtain an electrical signal of 0-1.
  • the electrical signal is the non-linear display light luminance NL_Yd in a non-linear space;
  • the curve can also use other non-linear conversion curves.
  • the brightness mapping curve TM_Curve can be TM_Curve in step 500; the non-linear display brightness
  • the signal NL_Yd is the brightness signal before the brightness mapping, which can also be said to be a non-linear display light brightness signal of the source image signal, and NL_Yt is the non-linear display brightness signal after the brightness mapping, which can also be said to be compatible with the display capability of the display device.
  • a non-linear display light brightness signal NL_Yd is input, and a corresponding brightness brightness non-linear brightness NL_Yt is obtained based on the calculation formula of the brightness map curve.
  • a one-dimensional lookup table LUT is selected to implement the brightness mapping curve.
  • a lookup table linear interpolation method may be adopted in actual mapping.
  • Linear interpolation is an interpolation method for one-dimensional lookup table LUTs, which estimates values based on two data points adjacent to the left and right of the point in the one-dimensional data sequence that needs to be interpolated.
  • the non-linear electrical signal is converted to a linear display light luminance signal in the range [0,10000] nits. It should be understood that NL_Yt can be converted to Yt by using a curve other than the PQ EOTF curve.
  • the ratio of K to the linear display luminance signal Yt after the luminance mapping and the linear display luminance signal Yd before the luminance mapping can be used to measure the change of the display luminance before and after the luminance mapping;
  • Rt, Gt, and Bt are the red, green, and blue components of the linear display light signal after the luminance mapping process, respectively.
  • BLoffset is the black level of the display device. In theory, when the screen displays black, it corresponds to The brightness should be 0, but in actual application, when the screen displays black, the corresponding brightness value is not 0, but a relatively small brightness value, that is, the black level is the minimum brightness that the display device can display; in this step BLoffset is the display brightness in linear space. It should be understood that the BLoffset of each display device may be different, and the BLoffset can be obtained by measuring the display device.
  • the processed linear display light signal RtGtBt After obtaining the processed linear display light signal RtGtBt, according to the color space of the actual display device, color space conversion is performed on the linear display light signal RtGtBt, and the RtGtBt signal is converted to the color space of the display device to obtain the display brightness range with the display device.
  • the processed video signal that matches the color space; for example, if the color space of the display device is sRGB space, it can be displayed directly without color space conversion; if the color space of the display device is YUV, RtGtBt The signal is converted into a video signal YUV1 in YUV space.
  • the brightness signal of the display light signal is converted to a non-linear space, and the brightness of the display brightness is mapped in the non-linear space, so that the display brightness range of the HDR video signal can be reasonably mapped to a display device capable of displaying In the brightness range, the contrast, brightness, and detail performance of the picture are improved. Since the brightness mapping is performed in a non-linear space, the errors introduced by the brightness mapping are evenly distributed, which has a small impact on the final display effect of the video signal, especially at low brightness. In the case of display, the display brightness distribution after mapping is reasonable, and the screen display will not be dark.
  • the embodiment of the present application considers the influence of the black level of the display device on the brightness mapping curve, and retains the brightness details of the low brightness portion. Further, the HDR signal brightness processing method provided in the embodiment of the present application can convert an HDR signal into an SDR signal, thereby improving the compatibility of the SDR display device with the HDR signal.
  • FIG. 9 a flowchart of a method for calculating a brightness mapping curve TM_Curve according to an embodiment of the present application. It should be understood that the method shown in FIG. 9 can be used to calculate the brightness mapping curve TM_Curve in step 500, and the brightness mapping curve TM_Curve in 500 can also be calculated by other methods, and is not limited to the method shown in FIG. 9.
  • the method for calculating the brightness mapping curve TM_Curve may include:
  • Source maximum brightness MaxSrcLuma, source minimum brightness MinSrcLuma, display device maximum brightness MaxDispLuma, display device minimum brightness MinDispLuma, units are nits; source maximum / minimum brightness can be determined according to actual conditions and experience, or metadata information carried from HDR signals To obtain the maximum / minimum brightness of the main control monitor as the maximum / minimum brightness of the source; the maximum / minimum brightness of the display device is measured according to the actual display device. Optionally, the brightness of the actual display device cannot be measured in special application scenarios. The value can be set based on experience.
  • the same nonlinear space PQ EOTF -1 curve as in 503 may be selected, and the maximum nonlinear brightness maxSL, the minimum nonlinear brightness minSL, the maximum nonlinear brightness maxDL, Non-linear display minimum brightness minDL;
  • step 901 may be considered as mapping a luminance signal in a linear space with a luminance range of [0,10000] nits to an electrical signal in a range of [0,1].
  • KP is the inflection point of the brightness mapping curve, KneePoint. Brightness values lower than KneePoint are not compressed, and brightness values higher than KneePoint are compressed; for example, KP is selected from (minDL, maxDL), and KP can be adjusted according to the actual effect. In an optional case, the selection of KP is related to the difference between the source brightness and the display brightness of the display device. When the source brightness is less than or equal to the display brightness, there is no need to compress the source brightness.
  • KP MaxSL; when the source brightness is much larger than the display brightness, the brightness interval to be compressed is larger, and KP chooses a smaller value; when the source brightness is greater than the display brightness and the difference between the two is not large, the value of KP can be taken as a larger value.
  • Hermite interpolation uses cubic double Hermite interpolation. It should be understood that the interpolation method can also use other interpolation methods. The formula is as follows:
  • x KP
  • x1 maxSL
  • y0 KP
  • y1 maxDL
  • y0 ′ 1
  • y1 ′ 0.
  • the display device Because the display device has a minimum brightness, even if a pure black signal is given to the display device, it will be displayed as the minimum brightness value on the display device. This minimum brightness is called the black level. Therefore, the display brightness needs to be raised to a certain level by BlackLevelLift, which is abbreviated as BLL to protect the detail brightness whose brightness is less than the threshold level.
  • the threshold level boost BLL value is based on the input of the brightness mapping curve TM_Curve. Calculation:
  • the black level rise BLL here is the brightness value in the non-linear space
  • the BLoffset in the aforementioned 507 is the brightness value in the linear space to ensure that the brightness mapping curve TM_Curve can map the minimum brightness of the source to the minimum brightness of the screen
  • the BLL increase TM_Curve has a small change, and the BLL needs to be normalized to obtain the normalized BLLnorm; the normalization result is as follows:
  • BLLnorm MAX ((BLL-norm_y1) / (norm_y0-norm_y1) * (minDL-maxSL), 0) (35)
  • norm_x0 minSL
  • norm_y0 minDL * (1-minSL) ⁇ n;
  • norm_x1 KP
  • norm_y1 minDL * (1-KP) ⁇ n;
  • the brightness mapping curve TM_Curve outputs e2 e1 + BLLnorm, where BLLnorm is the brightness value improvement caused by the black level after normalization;
  • TM_Curve_x (x0, x1, ... xn) is the abscissa of n sampling points, that is, the input of the curve, represents the brightness signal before the brightness mapping;
  • TM_Curve_y (y0, y1, ... yn) is the ordinate of n sampling points, that is, the output of the curve, which represents the brightness signal after brightness mapping;
  • mapping curve TM_Curve can be expressed in other ways (such as formulating) according to requirements; for details, refer to the description of TM_Curve in 500.
  • the brightness mapping curve is a relative value mapping curve.
  • the input of the mapping curve is [0,1] for [minSL, maxSL]
  • the output of the mapping curve is [0, 1] represents [minDL, maxDL]; in the embodiment of the present application, the brightness mapping curve directly calculates the output brightness based on the input brightness.
  • the brightness mapping curve is an absolute value mapping curve.
  • the mapping curve input [0,1] represents [0,10000]. ] nits
  • the mapping curve output [0,1] means [0,10000] nits.
  • the traditional brightness mapping curve is divided into three parts by selecting two thresholds.
  • the low-luminance part that is, the brightness part below the first threshold
  • the high-luminance part that is, the brightness part that is greater than the first threshold
  • Compression, and the high-brightness part is further divided into two parts, in which the luminance part that is greater than the first threshold value and less than the second threshold value is flexibly compressed by a curve that is obtained by fitting the first threshold value and the second threshold value.
  • the part greater than the second brightness inflection point is subjected to the second compression, that is, the brightness greater than the second threshold is mapped to the second threshold; segmentation mapping of the brightness, fully taking into account the individual brightness Characteristics, retain brightness details as much as possible, and improve the rationality of brightness mapping.
  • the brightness mapping curve takes into account the influence of the black level of the display device on the brightness mapping curve, and retains the brightness details of the low brightness portion.
  • FIG. 10 it is a flowchart of another method for processing brightness of an HDR video signal according to an embodiment of the present application.
  • the source video signal input in the embodiment of the present application is a scene light signal HDR and an HLG signal.
  • the method may include:
  • the input video signal in the embodiment of the present application is HDR in YUV space, and the HLG signal YUV0.
  • a nonlinear scene light signal R'sG'sB's in RGB color space is obtained; further, the nonlinearity
  • the scene light signal R'sG'sB's is converted into a linear scene light signal RsGsBs.
  • the nonlinear scene light signal R'sG'sB's is converted into a linear scene light signal RsGsBs based on the HLG OETF-1 curve.
  • the conversion from YUV to RGB is considered to be a color space conversion, and the non-linear RGB to linear RGB signals are not converted in color space.
  • the non-linear RGB signals and the linear RGB signals Both belong to the RGB color space.
  • the input video signal is converted from a YUV signal to a non-linear scene light signal R'sG'sB's into a color space conversion and the non-linear scene light signal R'dG
  • the conversion of 'sB's to the linear scene light signal RsGsBs can also be considered as a color space conversion.
  • the nonlinear RGB scene light signal and the linear RGB scene light signal belong to different color spaces.
  • Ys is the linear scene brightness signal
  • the scene brightness signal is the brightness component of the scene light signal.
  • the choice of the parameters cr, cg, and cb is related to the color gamut of the linear scene light signal RsGsBs.
  • the color gamut of the light signal of the linear scene is different, the corresponding calculation parameters will also be different. Therefore, the brightness needs to be calculated according to where the light signal is located.
  • the HLG OOTF defined in ITU BT.2100 is used to convert the linear scene light signal HLG to a linear display light signal for display. This method calculates the maximum brightness and minimum brightness of the actual display device. "System Gamma" related parameters convert linear scene light signals into linear display light signals within the dynamic range of the display device.
  • HLG OOTF is defined as follows:
  • L W and L B are the maximum brightness and the minimum brightness of the display device, respectively.
  • the scene brightness Y S is obtained, it is converted into display brightness, and brightness mapping processing is performed on the display brightness, or tone mapping processing is performed.
  • the brightness obtained based on the scene light signal is a scene light brightness signal, and the scene light brightness signal needs to be converted into a display light brightness signal.
  • This step is an exemplary implementation of the brightness signal type conversion.
  • Linear scene brightness Ys is converted to linear display brightness Yd;
  • the maximum brightness of the display signal LW 1000 nits is set; LW may also set other values.
  • Converting scene brightness to display brightness improves compatibility with the processing of scene signals and display signals.
  • step 603 Please refer to the description of step 603, which will not be repeated here.
  • step 604 Same as step 604, please refer to the description of step 604, which will not be repeated here.
  • step 605 Same as step 605, please refer to the description in step 605, which will not be repeated here.
  • K Yt / Ys, that is, K represents the ratio of the luminance signal Yt after luminance mapping in the linear display space to the scene luminance signal Ys before luminance mapping;
  • Rt, Gt, and Bt are the red, green, and blue components of the linear display light signal after the luminance mapping process, respectively;
  • BLoffset is the black level of the display device. For details, refer to the description in step 507.
  • the processed linear display light signal RtGtBt After obtaining the processed linear display light signal RtGtBt, according to the color space of the actual display device, color space conversion is performed on the linear display light signal RtGtBt, and the RtGtBt signal is converted to the color space of the display device to obtain the display brightness range with the display device.
  • the processed video signal that matches the color space; for example, if the color space of the display device is sRGB space, it can be displayed directly without color space conversion; if the color space of the display device is YUV, RtGtBt The signal is converted into a video signal YUV1 in YUV space.
  • the luminance signal of the scene light signal is converted into a display light luminance signal, and the linear display light luminance signal is converted into a non-linear space, and the luminance of the display brightness is mapped on the non-linear space.
  • the display brightness range of the video signal is reasonably mapped to the brightness range that the display device can display to improve the contrast, brightness, and detail performance of the picture; after the HLG scene signal is converted into a linear display light brightness signal, the display is not directly performed, but instead The linear display brightness signal is converted into a non-linear space, and the brightness mapping is performed in the non-linear space.
  • the embodiment of the present application considers the influence of the black level of the display device on the brightness mapping curve, and keeps the low Brightness details in the brightness section. Further, the HDR signal brightness processing method provided in the embodiment of the present application can convert an HDR signal into an SDR signal, thereby improving the compatibility of the SDR display device with the HDR signal.
  • FIG. 6, FIG. 9, and FIG. 10 describe the methods in the form of steps. Although the method sequence is shown in the method flowcharts 6, 9, and 10, In some cases, the steps described may be performed in a different order than described here.
  • a device for video signal processing includes a brightness obtaining unit 1301, a first conversion unit 1302, a brightness mapping unit 1303, a second conversion unit 1304, and a gain calculation unit. 1305 and a display signal acquisition unit 1306.
  • the device may further include a compensation unit 1307 and a color space conversion unit 1308.
  • the brightness obtaining unit 1301 is configured to obtain a first linear brightness signal, where the first linear brightness signal is obtained based on a first linear red-green-blue RGB signal corresponding to a video signal to be processed. For details, refer to the description in step 501, and details are not described herein again.
  • the brightness obtaining unit 1301 is specifically configured to perform color space conversion on the PQ signal to obtain a first nonlinear RGB signal; according to the PQ electro-optical transfer function, The first non-linear RGB signal is converted into the first linear RGB signal; calculation is performed based on each primary color signal of the first linear RGB signal to obtain the first linear luminance signal.
  • the brightness acquisition unit is specifically configured to: perform color space conversion on the HLG signal to obtain a second non-linear RGB signal;
  • the second non-linear RGB signal is converted into a second linear RGB signal; calculation is performed based on each primary color signal of the second linear RGB signal to obtain a third linear luminance signal; and the third linear luminance signal is subjected to luminance signal type conversion to The first linear luminance signal is obtained.
  • a first conversion unit 1302 configured to convert the first linear luminance signal into a first non-linear luminance signal
  • the first conversion unit 1302 may also be used to complete steps 603 and 1004;
  • a brightness mapping unit 1303, configured to perform segmented brightness mapping on the first nonlinear brightness signal to obtain a second nonlinear brightness signal;
  • the brightness mapping unit 1303 may also be used to complete steps 604 and 1005. It should be understood that the brightness mapping unit may call a look-up table, brightness mapping curve, or brightness mapping formula stored in the memory to complete the brightness mapping of the video signal to be processed.
  • the brightness mapping curve please refer to the description of the embodiment corresponding to 600 and FIG. 9.
  • the second conversion unit 1304 may also be used to complete steps 605 and 1006;
  • a gain calculation unit 1305, configured to calculate a brightness gain of the second linear brightness signal and the first linear brightness signal
  • the gain calculation unit 1305 may also be used to complete steps 606 and 1007;
  • a display signal obtaining unit 1306, configured to obtain an RGB display signal corresponding to the video signal to be processed based on a product of the brightness gain and the first linear RGB signal;
  • a compensation unit 1307 is configured to superimpose a black level on each primary color value of the RGB display signal and increase BlackLevelLift to obtain a processed RGB display signal, where the BlackLevelLift is a minimum value of the display brightness of the display device;
  • the display signal acquisition unit 1306 and the compensation unit 1307 may be used together to complete steps 607 and 1008;
  • the color space conversion unit 1308 is configured to perform color space conversion on the processed RGB display signal to obtain a target display signal.
  • the color format of the target display signal is the same as the color format corresponding to the display device. It should be understood that if the color format corresponding to the display device is RGB, the display can be performed directly, and no color space conversion is required.
  • the following is a specific implementation manner of video signal processing provided by the present application.
  • the embodiment of the application converts the HDR HLG signal into an SDR signal to adapt to the SDR TV.
  • saturation mapping is performed first, and then brightness mapping and color gamut mapping are performed.
  • the processing order of saturation mapping, brightness mapping, and color gamut mapping may be reversed. This is not limited.
  • Y ⁇ sCbsCrs be a 4: 4: 4 YCbCr non-linear video signal restored by the terminal after AVS2 decoding and reconstruction and chroma upsampling.
  • Each component is a 10-bit digitally encoded value.
  • Ynorm should clip to [0,1]
  • f sm () is a saturation mapping curve, which is calculated according to the brightness mapping curve f tm (), and the calculation steps are:
  • L is the input linear brightness
  • the unit is nit
  • the result of f tm (L) is the linear brightness, the unit is nit;
  • e is the normalized HLG signal brightness
  • f tmHLG (e) is the normalized HLG signal brightness
  • the saturation mapping curve input e, f sm (e) is the saturation mapping gain on the HLG space
  • the YiCbiCri signal is a 10-bit limited range digital coded value, where the Yi value should be in the [64,940] interval, and the Cbi, Cri value should be in the [64,960] interval.
  • Y ⁇ s Cb s Cr s signal is a 10-bit digital code value limits, obtained through the process R ⁇ s G ⁇ s B ⁇ s nonlinear primary values are floating-point numerical clip to be [0,1] Interval.
  • Equation E s s B s represents the signal R s G value of any of a linear primary color component has a value in the interval [0,1];
  • E ⁇ s refers R ⁇ s G ⁇ s B ⁇ s signal to any one of The non-linear primary color value of the component.
  • the function HLG_OETF -1 () is defined according to ITU BT.2100 as follows:
  • the linear brightness Y s is calculated as follows:
  • Y s is a real number whose value is in the interval [0,1].
  • hmt (x) 0.4064 ⁇ ⁇ 0 (x) + 0.5791 ⁇ ⁇ 1 (x) + ⁇ 0 (x)
  • Y t is a real number, and its value should be clipped to the interval [0,200].
  • Es represents any component in the R s G s B s signal
  • E tm represents any component in the R tm G tm B tm signal.
  • the RtGtBt obtained through this process is a floating-point linear primary color value, and the value should be clip to [0,200].
  • ⁇ in this embodiment may be 2.2 or 2.4, or other values.
  • the value of ⁇ may be selected according to actual conditions, which is not limited in the embodiment of the present application.
  • R ⁇ tG ⁇ tB ⁇ t is a non-linear primary color value, and the value is in the interval [0,1].
  • the Y ⁇ tCbtCrt signal obtained after this processing is a 10-bit limited range digital coded value, where the Y ⁇ t value should be in the [64,940] interval, and the Cbt, Crt values should be in the [64,960] interval.
  • a process flowchart of an HDR terminal technical solution provided in the present application shows a technical framework and a related scope in the entire video end-to-end system.
  • the technical framework of the end-to-end system provided by the embodiment of the present application can convert HDR HLG video signals into SDR signals, thereby adapting to SDR TV, and can also convert HDR HLG signals into HDR PQ signals, adapting HDR PQ TV for display; optionally, HDR PQ signals can also be converted to SDR signals for playback on SDR TV; optionally, dynamic metadata of video signals can be discarded, and only static metadata is retained, which is suitable for Play with HDR PQ TV.
  • the technique will be a second-generation audio frame video signal and the video coding standard HDR dynamic metadata or static metadata (2 nd Audio Video coding Standard, AVS2) encoded into code AVS2 streamed to the terminal, optionally,
  • HDR dynamic metadata or static metadata (2 nd Audio Video coding Standard, AVS2) encoded into code AVS2 streamed to the terminal
  • HEVC High Efficiency Video Coding
  • the HDR input signal source supported in the embodiment of the present application is an AVS2 code stream received at a terminal.
  • an integer YCbCr color difference signal (hereinafter referred to as YCC signal for short, which is not described in this article) in the format of 10bit 4: 2: 0 and related HDR static or dynamic metadata are obtained.
  • YCC and HDR signals may be in several formats listed in Table 1:
  • the TVs that may be connected to the terminal have very different support capabilities for HDR signals. It is often difficult for older TV models to support new signal formats. To this end, the decoding terminal needs to perform compatible adaptation according to the difference in the TV's ability to support HDR signals.
  • the TVs that terminal devices may need to interface with are classified into the following categories, as shown in Table 3:
  • the embodiment of the present application performs corresponding signal conversion processing according to different HDR signal formats received by the terminal and different HDR signal support capabilities of the docking TV, so as to achieve compatible adaptation between the signal and the TV.
  • YsCbsCrs be a 4: 4: 4 YCbCr non-linear video signal restored by the terminal after AVS2 decoding and reconstruction and chroma upsampling.
  • Each component is a 10-bit digitally encoded value.
  • the YsCbsCrs signal is a 10-bit limited range digital coded value.
  • the R ⁇ sG ⁇ sB ⁇ s obtained through this process is a floating-point non-linear primary color value, and the value should be clip to the [0,1] interval.
  • the input HDR HLG video signal is a video signal in YCC space. It can also be said that HLG is a scene light signal.
  • the HLG video signal is converted to the RGB color space to obtain a nonlinear scene light signal R ⁇ sG ⁇ sB. ⁇ s.
  • the brightness mapping process is completed in the RGB color space. It should be understood that the video signal obtained by the color space conversion is a non-linear video signal. If the input is an HLG signal, the color space conversion is a non-linear scene light signal.
  • Equation E s s B s represents a signal of any one of R s G component;
  • E ⁇ s refers R ⁇ s G ⁇ s B ⁇ s signal to any one component.
  • the function HLG_OETF -1 () is defined according to ITU BT.2100 as follows:
  • a first linear display luminance signal is obtained based on the linear video signal.
  • the first linear display luminance signal is a linear display luminance signal.
  • the linear video signal is a linear scene light signal due to the input HDR HLG signal
  • the brightness signal Ys obtained based on the linear scene light signal is a scene brightness signal.
  • the linear brightness Ys is calculated as follows:
  • the input signal is HDR, HLG signal, and scene light signal
  • the Ys calculated in the above steps is the linear scene brightness
  • the input of the brightness mapping curve is the display brightness, so it needs to be converted into a display before performing the brightness mapping.
  • Lightness signal Yd is the input signal.
  • the brightness mapping is performed in a non-linear space, so the brightness mapping input is a non-linear display brightness, so the linear display brightness needs to be converted into a non-linear space to obtain the non-linear display brightness Y dPQ ;
  • the signal obtained after the luminance mapping is a non-linear display light luminance signal.
  • the non-linear display light luminance signal can be converted into a linear display light luminance signal Yt through a PQ EOTF curve.
  • Yt is a real number, and its value is in the interval [0,100].
  • ⁇ in this embodiment may be 2.2 or 2.4, or other values.
  • the value of ⁇ may be selected according to actual conditions. This embodiment of the present application does not limit this. .
  • E s represents any component in the R s G s B s signal
  • E tm represents any component in the R tm G tm B tm signal.
  • the color gamut mapping may be:
  • the function EOTF -1 () may be the inverse function of BT.1886EOTF or the inverse function of BT.2100PQ EOTF curve.
  • E ⁇ t (E t / 100) 1 / 2.4 .
  • R ⁇ tG ⁇ tB ⁇ t is a floating-point non-linear primary color value, and the value is in the interval [0,1].
  • the YtCbtCrt signal obtained after this processing is a 10-bit limited range digitally encoded value.
  • the YoCboCro signal is a 10-bit limited range digital coded value.
  • This process is suitable for compatible adaptation processing of HDR HLG signals and HDR PQ signals.
  • the parameters of each step of the above processing flow are different. It should be understood that, in the embodiment of the present application, the video signal is first subjected to brightness processing, then color gamut processing, and then saturation processing. In an optional case, color gamut processing, brightness processing, and saturation processing may be performed first.
  • this embodiment proposes to perform compatible adaptation in the manner listed in Table 4.
  • this embodiment also converts the HDR HLG signal to the BT.709 color gamut and sends it to the type 2 TV.
  • This part of the processing is a link of HLG signal to SDR signal compatible adaptation processing. Since this processing method has been given a conceptual introduction in the BT.2407 report, this section refers to the content of the ITU report for informational explanation.
  • the conversion of BT.2020 wide color gamut signals to BT.709 signals can be achieved by a method based on linear matrix conversion. In addition to hard-clip the output signal, this method is completely the inverse process of ITU standard BT.2087.
  • each component signal through a linear transfer function to achieve conversion signal (E R E G E B)
  • the conversion function may be an HLG EOTF function.
  • BT709 linear RGB signals (ER R G E B ) are to be used in BT709 display devices, and should be converted to BT709 nonlinear RGB signals (E ⁇ R E ⁇ G E ⁇ B ).
  • ⁇ in this embodiment may be 2.2 or 2.4, or other values.
  • the value of ⁇ may be selected according to actual conditions, which is not limited in the embodiment of the present application.
  • the first reference peak brightness Lw from HLG to PQ signal is 1000nit, and the black level Lb is 0.
  • the same PQ image as the HLG image can be generated in the color volume within 1000nit using the process shown in Figure 16:
  • 1000nit HLG source signal can generate linear scene light signal after HLG OETF inverse function
  • the linear scene light signal can generate a linear display light signal through the OOTF function of HLG;
  • the linear display light signal can generate 1000nit PQ display light signal through the EOTF inverse function of PQ;
  • This processing flow involves the process of converting HDR HLG signals into HDR PQ signals and displaying them on TV.
  • YsCbsCrs be a 4: 4: 4 YCbCr non-linear video signal restored by the terminal after AVS2 decoding and reconstruction and chroma upsampling.
  • Each component is a 10-bit digitally encoded value.
  • the YsCbsCrs signal is a 10-bit limited range digital coded value.
  • the R ⁇ sG ⁇ sB ⁇ s obtained through this process is a floating-point non-linear primary color value, and the value should be clip to the [0,1] interval.
  • the input HDR HLG video signal is a video signal in YCC space. It can also be said that HLG is a scene light signal.
  • the HLG video signal is converted to the RGB color space to obtain a nonlinear scene light signal R ⁇ sG ⁇ sB. ⁇ s.
  • the brightness mapping process is completed in the RGB color space. It should be understood that the video signal obtained by the color space conversion is a non-linear video signal. If the input is an HLG signal, the color space conversion is a non-linear scene light signal.
  • the non-linear scene light signal R ⁇ sG ⁇ sB ⁇ s is converted into a linear scene light signal RsGsBs.
  • the non-linear scene light signal R ⁇ sG ⁇ sB ⁇ s can be based on the HLG photoelectric transfer inverse function. Converted into linear scene light signals RsGsBs.
  • HLG_OETF -1 () is defined according to ITU BT.2100 as follows:
  • a first linear display luminance signal is obtained based on the linear video signal.
  • the first linear display luminance signal is a linear display luminance signal.
  • the linear video signal is a linear scene light signal due to the input HDR HLG signal
  • the brightness signal Ys obtained based on the linear scene light signal is the scene brightness signal. It needs to be converted into display light before performing brightness mapping. Brightness signal Yd.
  • the linear brightness Ys is calculated as follows:
  • Es represents any component in the RsGsBs signal
  • E tm represents any component in the R tm G tm B tm signal.
  • R ⁇ tG ⁇ tB ⁇ t is a floating-point non-linear primary color value, and the value is in the interval [0,1].
  • the YtCbtCrt signal obtained after this processing is a 10-bit limited range digitally encoded value.
  • the terminal may complete the frame rate adjustment, the bit width adjustment, and the 4: 4: 4 to 4: 2: 2/4: 2 according to the frame rate, bit width, and chroma downsampling method determined by the HDR and PQ TV. : 0 Downsampling and other subsequent processing can transmit the HDR PQ signal generated by the conversion to the HDR PQ TV.
  • this embodiment proposes to perform compatible adaptation in the manner listed in Table 5.
  • the terminal can complete the frame rate adjustment, bit width adjustment, and 4: 4: 4 to 4: 2: 2/4: 2 according to the frame rate, bit width, and chroma downsampling method determined by the connected SDR TV. Subsequent processing such as 0 downsampling can transmit the SDR signal generated by the conversion to the SDR TV.
  • HDR PQ signals have HDMI 2.0A or higher HDMI interface capabilities, and have more mature HDR PQ signal processing capabilities
  • this embodiment suggests that when receiving HDR PQ signals, the HDR PQ signal and The static metadata is directly output to the HDR, PQ, and TV through the HDMI interface of HDMI 2.0A and above.
  • the TV then completes the subsequent display processing of the HDR and PQ signals.
  • the dynamic metadata it cannot be passed to the TV due to the limitation of the HDMI interface, and it should be discarded after decoding.
  • the specific method is to continuously broadcast the image to be evaluated and the reference image to the observer according to the test schemes described later, and then leave a certain time interval for the observer to score after the playback, and finally average all the given scores as the
  • the evaluation value of the sequence is the evaluation value of the image to be evaluated.
  • HDR and HLG signals adapt to SDR and TV
  • Test purpose Through comparative tests, it is shown whether the image adaptation conversion from HLG to SDR can provide beneficial image effects when the HDR and HLG signals are sent to the SDR and TV.
  • Figure 17 shows a schematic diagram of a test network. among them:
  • DUT2 HLG to SDR BT2020
  • BenchMark1 HLG watched on SDR BT709
  • BenchMark1 HLG watching in HDR, HLG, TV in HLG, BT2020 mode
  • HDR HLG signal adapts HDR PQ TV
  • FIG. 18 it is a schematic diagram of a networking manner of another test solution according to an embodiment of the present application.
  • Benchmark1 HLG watching in SDR and BT709 mode on HDR PQ TV
  • Benchmark2 HLG watching on HDR PQ TV in HLG BT2020 mode
  • HDR end-to-end system For PQ curve HDR video, the HDR end-to-end system is shown in Figure 19.
  • program production PQ curve HDR video and static metadata are obtained, and the HDR video production parameters meet the requirements of GY / T 315-2018 standard.
  • HDR pre-processing realizes the extraction of dynamic metadata, obtains HDR video and metadata for encoding and transmission, and encodes and encapsulates the AVS2 before transmitting it on the network.
  • the AVS2 decoder decodes the HDR video and metadata.
  • HDR video and metadata reconstruction is used to obtain SDR video for display; for HDR display terminals, if the terminal display capability is the same as the HDR video produced and transmitted, the HDR display is directly performed; if the terminal display capability and produced transmission The brightness of the HDR video is different, and then the HDR video and metadata are adapted to be displayed according to the display capability of the terminal.
  • HLG curve HDR video when the program production uses the maximum brightness of 1000cd / m2, the HDR end-to-end system is shown in Figure 20. Through program production, HLG curve HDR video is obtained, and the production parameters of HDR video are in compliance with GY / T315-2018 standards. After using AVS2 encoding for HDR video, it is transmitted on the network. At the receiving end, the AVS2 decoder decodes the HDR video and displays it directly on the SDR and HDR terminals.
  • HLG curve HDR video when the highest brightness of the program production is not 1000cd / m2, the HDR end-to-end system is shown in Figure 21.
  • HLG curve HDR video and static metadata are obtained.
  • HDR video production parameters meet the requirements of GY / T315-2018 standard.
  • HDR video and static metadata are encoded and encapsulated in AVS2 and transmitted over the network.
  • the AVS2 decoder obtains HDR video after decoding
  • the AVS2 decoder obtains HDR video and static metadata after decoding.
  • SDR display terminals direct display; for HDR display terminals, static metadata can be used to display the adjusted gamma value using the method specified in Table 5 of GY / T315-2018.
  • Metadata information metadata_info contains dynamic metadata that can be used to reconstruct HDR and images when combined with the associated transmitted image.
  • hdr_characteristics () contains the HDR image signal characteristics, that is, the identification of the HDR image color space, and the identification for adjusting the primary color of the HDR image monitor.
  • This variable represents the primary color and the reference white coordinates that conform to the HDR image color space defined in CIE1931.
  • This variable represents the primary and reference white coordinates of the main monitor color space of the adjusted HDR image as defined by CIE1931.
  • This variable defines the nominal maximum display brightness of the main monitor that adjusts the HDR image, measured in candela per square meter (cd / m2) and rounded to an integer multiple of 50 cd / m2.
  • This variable defines the nominal minimum display brightness of the main monitor that adjusts the HDR image. In 0.0001cd / m2.
  • hdrDisplayMinLuminance should be less than hdrDisplayMaxLuminance. If the value of the variable is unknown, it is recommended to set it to 0.
  • sdr_characteristics contains the characteristics of the SDR image signal, that is, the color space identification of the SDR image, and the nominal maximum and minimum brightness values used to adjust the SDR image main monitor.
  • This variable defines the nominal maximum display brightness of the main monitor used to adjust the SDR image, in units of 1cd / m2, and rounded to an integer multiple of 50cd / m2.
  • luminance_mapping_variables contains luminance mapping variables used to build the lookup table lutMapY.
  • This variable is used in the first step of calculating the signal gain during the reconstruction of the luminance mapping curve, and represents the black level offset eliminated during the chrominance volume reconstruction.
  • the value of this variable should be in the range [0,1] and be an integer multiple of (1 ⁇ 255).
  • the set-top box learns the display capabilities of the TV.
  • the set-top box analyzes the received program stream and obtains the information that the program is SDR program / HDR program, maximum brightness, minimum brightness, etc. If it matches the display capability of the TV, it decodes the audio and video and sends it to the TV for display via HDMI. ; If it does not match the TV ’s actual capabilities, decode the audio and video and obtain the program signal adapted to the TV ’s display capabilities through display adaptation, and send it to the TV via HDMI for display.
  • the all-in-one decodes the received program code stream and adjusts and displays it according to the actual display capability of the TV.
  • This embodiment of the present application describes a method for transmitting HDR metadata in an HDMI interface.
  • the HDMI 2.0a specification specifies how to transmit HDR static metadata in the HDMI interface
  • the HDMI 2.1 specification specifies how to transmit HDR dynamic data in the HDMI interface. Metadata.
  • the HDMI 2.0a standard adopts the CEA-861.3-2014 specification for HDR static metadata transmission.
  • upstream source processing equipment such as a set-top box
  • the HDR static metadata can be in the CEA-861.3-2014 interface It is transmitted to a receiver (such as a television) that can receive and process HDR static metadata.
  • This appendix explains the HDR static metadata data block "HDR Static Metadata Data Block” (C.2.2, corresponding to CEA-861.3-2014 specification 4.2) and the dynamic range main supervisor information frame "Dynamic MetaMetadata" in CEA-861.3-2014.
  • Mastering InfoFrame “(C.2.3, corresponding to 3.2 of CEA-861.3-2014 specification) information transmission and mapping.
  • the HDR static metadata data block "HDR Static Metadata Data Block” is used to transmit the receiving end HDR static metadata support capability information to the source device.
  • the "ET_2" bit in the data block transmitted from the receiver to the source device is equal to 1, it means that the receiver supports the PQ EOTF curve specified by GY / T315-2018, and the "SM_0" bit is equal to 1, which means that it supports the static elements specified in this article. data.
  • the dynamic range master information frame "Dynamic Range and Mastering InfoFrame" is used by the source device to identify and transmit HDR dynamic metadata to the receiving device.
  • the source device uses the "EOTF” value of 2 to mark the EOTF of the transport stream as the PQ curve specified by GY / T315-2018, and "Static_MetaData_Descriptor_ID" is 0 to mark that the static metadata of this standard is carried in the user extension information, so that the transmission complies with Static metadata specified in this standard.
  • the HDMI 2.1 standard uses the CTA-861-G-2016 specification for HDR dynamic metadata transmission.
  • CTA-861-G2016 specifies how to carry HDR metadata in the CE terminal interface.
  • the upstream source processing equipment such as a set-top box
  • the HDR metadata can be transmitted in the CTA-861-G-2016 interface to a receiver (such as a television) that can receive and process the HDR metadata.
  • This appendix explains the HDR dynamic metadata data block "HDR Dynamic Metadata Data Block” (C.3.2, corresponding to 7.5.14 of the CTA-861-G-2016 specification) and HDR dynamic metadata in CTA-861-G-2016.
  • Information transmission and mapping of the extended information frame "HDR Dynamic Metadata Extended InfoFrame” (C.3.3, corresponding to the 6.10.1 of the CTA-861-G-2016 specification).
  • the HDR dynamic metadata data block "HDR Dynamic Metadata Data Block” is used to transmit the receiving end HDR dynamic metadata support capability information to the source device.
  • the value of the supported dynamic metadata type "Supported HDR Dynamic Metadata Type" in the data block transmitted from the receiving end to the source device is 0x0002, it means that the receiving end supports the dynamic metadata specified in this standard.
  • the support flag "Support Flags" bytes in the data block represent different HDR transmission modes. Its binary value consisting of 0 to 3 bits is greater than or equal to 1, and the fifth bit is equal to 1, indicating that it supports the HDR transmission mode of this standard.
  • HDR dynamic metadata extended information frame "HDR Dynamic Metadata Extended InfoFrame" is used by the source device to identify and transmit the HDR dynamic metadata to the receiving device.
  • the source device uses the extended information frame type "Extended InfoFrame Type" value of 0x0002 to mark that this standard's dynamic metadata is carried in the user's extended information, thereby transmitting dynamic metadata that complies with the specifications of this standard.
  • An embodiment of the present application further provides a computer-readable storage medium.
  • the computer-readable storage medium stores instructions, and when the computer-readable storage medium runs on the computer, the computer causes one or more steps in any one of the foregoing methods to be executed.
  • each component module of the above signal processing device is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in the computer-readable storage medium.
  • the embodiments of the present application further provide a computer program product containing instructions.
  • the technical solution of the present application is essentially a part that contributes to the existing technology or all or part of the technical solution may be a software product.
  • the computer software product is stored in a storage medium, and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) or a processor therein to execute the embodiments of the present application. All or part of the steps of the method. Please refer to the relevant description of the memory 302 for the type of the storage medium.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Processing Of Color Television Signals (AREA)
  • Control Of Indicators Other Than Cathode Ray Tubes (AREA)
  • Controls And Circuits For Display Device (AREA)
  • Image Processing (AREA)
  • Transforming Electric Information Into Light Information (AREA)

Abstract

本申请实施例公开了一种视频信号处理的方法，该方法包括：获得第一线性亮度信号，该第一线性亮度信号基于待处理视频信号对应的第一线性红绿蓝RGB信号获得；将该第一线性亮度信号转换为第一非线性亮度信号；对该第一非线性亮度信号进行分段亮度映射，以得到第二非线性亮度信号；将该第二非线性亮度信号转换为第二线性亮度信号；计算该第二线性亮度信号与该第一线性亮度信号的亮度增益；基于该亮度增益和该第一线性RGB信号的乘积，获得该待处理视频信号对应的RGB显示信号；该方法将亮度信号转换到非线性空间上进行分段亮度映射，改善了视频信号在显示设备上的显示效果。

Description

一种视频信号处理的方法及装置
本申请要求于2018年7月5日提交中国国家知识产权局、申请号为201810732833.5、申请名称为“一种高动态范围视频信号处理的方法及装置”的中国专利申请，以及2018年07月19日提交中国国家知识产权局、申请号为201810797968.X、申请名称为“一种视频信号处理的方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及多媒体通信领域，尤其涉及一种视频信号处理的方法及装置。
背景技术
高动态范围(High Dynamic Range，HDR)视频技术拓展了可显示的图像的亮度范围，因而能够记录较大的亮度范围信息，展现更多的亮部和暗部细节。HDR是近几年视频行业出现的热门技术，也是未来视频行业发展的方向。现实世界中人眼看到的真实画面动态范围较大。而传统的标准动态范围(Standard Dynamic Range，SDR)显示设备亮度低，动态范围较小，传统的SDR视频技术在相机捕获、制作编辑、编码的过程中，不断压缩捕获画面的动态范围，使其适合在SDR电视设备上显示；HDR视频亮度通常远大于SDR显示设备或大量其他现有的HDR显示设备的能够显示的亮度范围；因此HDR视频信号在现有的显示设备上显示时，需要根据显示设备能力对HDR视频信号亮度进行处理，使其匹配显示设备可显示的亮度范围，适合在现有设备上显示，在对视频信号进行亮度处理时，不合理的亮度处理导致HDR视频的显示效果不佳。
发明内容
本申请的实施例提供一种视频信号处理的方法及装置，改善了视频信号在显示设备上的显示效果。
本申请第一方面提供了一种视频信号处理的方法，该方法包括：获得第一线性亮度信号，该第一线性亮度信号基于待处理视频信号对应的第一线性红绿蓝RGB信号获得；将该第一线性亮度信号转换为第一非线性亮度信号；对该第一非线性亮度信号进行分段亮度映射，以得到第二非线性亮度信号；将该第二非线性亮度信号转换为第二线性亮度信号；计算该第二线性亮度信号与该第一线性亮度信号的亮度增益；基于该亮度增益和该第一线性RGB信号的乘积，获得该待处理视频信号对应的RGB显示信号。
本申请实施例将亮度信号转换到非线性空间上进行分段亮度映射，可以将视频信号的显示亮度范围合理映射到显示设备能够显示的亮度范围内，提升画面的对比度、亮度及细节表现，尤其在低亮度显示的情况下映射后的显示亮度分布合理、画面显示不会偏暗；由于亮度映射是在非线性空间中进行的，由亮度映射引入的误差均匀分布，对视频信号最终的显示效果影响较小，另外，由于待处理的视频信号的亮度范围较大，不同亮度区域的视频信号对视频显示效果的贡献度也有所区别，基于待处理的视频信号的亮度区间的特性采 用不同的亮度映射关系对待处理视频信号的亮度值进行分段映射，提升了亮度映射的灵活性和合理性。
在一种可能的实施方式中，该待处理视频信号为感知量化PQ信号，该获得第一线性亮度信号，包括：对该PQ信号进行色彩空间转换，以获得第一非线性RGB信号；根据PQ电光转移函数，将该第一非线性RGB信号转换为该第一线性RGB信号；基于该第一线性RGB信号的各基色信号进行计算，以获得该第一线性亮度信号。
在一种可能的实施方式中，该待处理视频信号为混合对数伽玛HLG信号，该获得第一线性亮度信号，包括：对该HLG信号进行色彩空间转换，以获得第二非线性RGB信号；根据HLG光电转移逆函数，将该第二非线性RGB信号转换为第二线性RGB信号；基于该第二线性RGB信号的各基色信号进行计算，以获得第三线性亮度信号；对该第三线性亮度信号进行亮度信号类型转换，以获得该第一线性亮度信号。
在本申请实施例中，基于HLG信号得到的亮度信号为场景光亮度信号，需要将场景光亮度信号转换为显示光亮度信号，且在转换为显示光亮度信号之后，不直接进行显示，而是将亮度信号转换到非线性空间中进行分段亮度映射，尽可能保留亮度细节，提升亮度映射的合理性，提升HLG信号的显示效果。
在一种可能的实施方式中，该第一线性亮度信号为线性显示光亮度信号，该第三亮度信号为线性场景光亮度信号。
在一种可能的实施方式中，在该获得该待处理视频信号对应的RGB显示信号之后，还包括：对该RGB显示信号进行色彩空间转换，以获得目标显示信号，其中，该目标显示信号的色彩格式和显示设备对应的色彩格式相同。
在一种可能的实施方式中，在该获得该待处理视频信号对应的RGB显示信号之后，还包括：对该RGB显示信号的各基色值叠加黑位电平提升BlackLevelLift，以获得处理后的RGB显示信号，该BlackLevelLift为该显示设备的显示亮度的最小值；对应的，该对该RGB显示信号进行色彩空间转换，包括：对该处理后的RGB显示信号进行色彩空间转换。
本申请实施例考虑了显示设备的黑位电平对亮度映射曲线的影响，保留了低亮度部分的亮度细节。
在一种可能的实施方式中，其特征在于，对该第一非线性亮度信号进行分段亮度映射，以得到第二非线性亮度信号，包括：确定第一阈值和第二阈值，该第一阈值小于该第二阈值；当该第一非线性亮度信号的亮度值小于或等于该第一阈值时，该第二非线性亮度信号的亮度值等于该第一非线性亮度信号的亮度值；当该第一非线性亮度信号的亮度值大于该第一阈值，且小于或等于该第二阈值时，该第二非线性亮度信号的亮度值基于以该第一非线性亮度信号的亮度值为自变量的拟合曲线获得；当该第一非线性亮度信号的亮度值大于该第二阈值时，该第二非线性亮度信号的亮度值等于该显示设备对应的最大非线性显示亮度值。
本申请实施例根据两个亮度值阈值将待处理视频信号分成三段，将小于第一阈值的部分作为第一段视频信号，其亮度映射后的亮度等于亮度映射前的亮度，即对低亮度部分的视频信号不压缩，能最完整的保留低亮度部分的图像细节；将高亮度部分分成两段，其中，大于第一阈值且小于第二阈值的部分基于拟合曲线进行压缩，尽可能保留该部分亮度的细节；大于第二阈值的部分都压缩到第二亮阈值上；对亮度进行分段映射，充分考虑了各个 亮度的特性，尽可能保留亮度细节，提升亮度映射的合理性。
在一种可能的实施方式中，该拟合曲线通过对该第一阈值和该第二阈值进行埃尔米特Hermite插值得到。
在一种可能的实施方式中，其特征在于，对该第一非线性亮度信号进行分段亮度映射，以得到第二非线性亮度信号，包括：使用以下分段函数对该第一非线性亮度信号进行该亮度映射：
Figure PCTCN2019090566-appb-000001
其中，hmt(x)＝y 0×α 0(x)+y 1×α 1(x)+y′ 0β 0(x)+y′ 1β 1(x)
Figure PCTCN2019090566-appb-000002
e为该第一非线性亮度信号，f tm(e)为该第二非线性亮度信号，KP1为该第一阈值，KP2为该第二阈值，maxDL为该显示设备的最大非线性显示亮度值，maxSL为最大非线性源亮度值，x 0＝KP1，x 1＝maxSL，y 0＝KP1，y 1＝maxDL，y 0′＝1，y 1′＝0。
在本申请实施例中，y0＝x0表示在第一阈值下的亮度值不压缩，采用线性映射；x1＝maxSL,y1＝maxDL表示把源的最大亮度映射到屏的最大亮度maxDL；x0＝KP,y0'＝1保证曲线在KP这个阈值斜率为1，x1＝maxSL,y1'＝0保证曲线在maxSL这个阈值斜率为0，这样可以保证拟合得到的曲线在两个阈值处是平滑的。
在一种可能的实施方式中，其特征在于，该确定第一阈值和第二阈值，包括：根据该第一非线性亮度信号的显示亮度范围和该显示设备的显示亮度范围的关系确定该第一阈值；将该第一非线性亮度信号的最大亮度值作为该第二阈值。
在本申请实施例中，亮度值阈值的选取与源亮度和显示设备的显示亮度的差值有关，当源亮度小于等于显示亮度时，无需对源亮度进行压缩，此时第一阈值等于第二阈值等于源信号的非线性最大亮度值；当源亮度远远大于显示亮度，需要压缩的亮度区间较大，第一阈值选取较小的值；当源亮度大于显示亮度且两者相差不大，则第一阈值可以取较大的值。
在一种可能的实施方式中，其特征在于，该对该第一非线性亮度信号进行分段亮度映射，以得到第二非线性亮度信号，包括：基于预设的该第一非线性亮度信号和该第二非线性亮度信号的亮度值的映射关系，确定与该第一非线性亮度信号的亮度值相对应的第二非线性亮度信号的亮度值。
在一种可能的实施方式中，其特征在于，该将该第一线性亮度信号转换为第一非线性亮度信号，包括：根据PQ电光转移逆函数，将该第一线性亮度信号转换为该第一非线性亮度信号；对应的，该将该第二非线性亮度信号转换为第二线性亮度信号，包括：根据PQ 电光转移函数，将该第二非线性亮度信号转换为该第二线性亮度信号。
本申请第二方面提供了一种视频信号处理的装置，其特征在于，该装置包括：亮度获取单元，用于获得第一线性亮度信号，该第一线性亮度信号基于待处理视频信号对应的第一线性红绿蓝RGB信号获得；第一转换单元，用于将该第一线性亮度信号转换为第一非线性亮度信号；亮度映射单元，用于对该第一非线性亮度信号进行分段亮度映射，以得到第二非线性亮度信号；第二转换单元，用于将该第二非线性亮度信号转换为第二线性亮度信号；增益计算单元，用于计算该第二线性亮度信号与该第一线性亮度信号的亮度增益；显示信号获取单元，用于基于该亮度增益和该第一线性RGB信号的乘积，获得该待处理视频信号对应的RGB显示信号。
在一种可能的实施方式中，该待处理视频信号为感知量化PQ信号，该亮度获取单元，具体用于：对该PQ信号进行色彩空间转换，以获得第一非线性RGB信号；根据PQ电光转移函数，将该第一非线性RGB信号转换为该第一线性RGB信号；基于该第一线性RGB信号的各基色信号进行计算，以获得该第一线性亮度信号。
在一种可能的实施方式中，该待处理视频信号为混合对数伽玛HLG信号，该亮度获取单元，具体用于：对该HLG信号进行色彩空间转换，以获得第二非线性RGB信号；根据HLG光电转移逆函数，将该第二非线性RGB信号转换为第二线性RGB信号；基于该第二线性RGB信号的各基色信号进行计算，以获得第三线性亮度信号；对该第三线性亮度信号进行亮度信号类型转换，以获得该第一线性亮度信号。
在一种可能的实施方式中，该装置还包括：色彩空间转换单元，用于：对该RGB显示信号进行色彩空间转换，以获得目标显示信号，其中，该目标显示信号的色彩格式和显示设备对应的色彩格式相同。
在一种可能的实施方式中，该装置还包括：补偿单元，用于：对该RGB显示信号的各基色值叠加黑位电平提升BlackLevelLift，以获得处理后的RGB显示信号，该BlackLevelLift为该显示设备的显示亮度的最小值；对应的，该色彩空间转换单元，具体用于：对该处理后的RGB显示信号进行色彩空间转换。
在一种可能的实施方式中，该亮度映射单元，具体用于：确定第一阈值和第二阈值，该第一阈值小于该第二阈值；当该第一非线性亮度信号的亮度值小于或等于该第一阈值时，该第二非线性亮度信号的亮度值等于该第一非线性亮度信号的亮度值；当该第一非线性亮度信号的亮度值大于该第一阈值，且小于或等于该第二阈值时，该第二非线性亮度信号的亮度值基于以该第一非线性亮度信号的亮度值为自变量的拟合曲线获得；当该第一非线性亮度信号的亮度值大于该第二阈值时，该第二非线性亮度信号的亮度值等于该显示设备对应的最大非线性显示亮度值。
在一种可能的实施方式中，该拟合曲线通过对该第一阈值和该第二阈值进行埃尔米特Hermite插值得到。
在一种可能的实施方式中，该亮度映射单元，具体用于：使用以下分段函数对该第一非线性亮度信号进行该亮度映射：
Figure PCTCN2019090566-appb-000003
其中，hmt(x)＝y 0×α 0(x)+y 1×α 1(x)+y′ 0β 0(x)+y′ 1β 1(x)
Figure PCTCN2019090566-appb-000004
e为该第一非线性亮度信号，f tm(e)为该第二非线性亮度信号，KP1为该第一阈值，KP2为该第二阈值，maxDL为该显示设备的最大非线性显示亮度值，maxSL为最大非线性源亮度值，x 0＝KP1，x 1＝maxSL，y 0＝KP1，y 1＝maxDL，y 0′＝1，y 1′＝0。
在一种可能的实施方式中，确定第一阈值和第二阈值，包括：根据该第一非线性亮度信号的显示亮度范围和该显示设备的显示亮度范围的关系确定该第一阈值；将该第一非线性亮度信号的最大亮度值作为该第二阈值。
在一种可能的实施方式中，该亮度映射单元，具体用于：基于预设的该第一非线性亮度信号和该第二非线性亮度信号的亮度值的映射关系，确定与该第一非线性亮度信号的亮度值相对应的第二非线性亮度信号的亮度值。
在一种可能的实施方式中，该第一转换单元，具体用于：根据PQ电光转移逆函数，将该第一线性亮度信号转换为该第一非线性亮度信号；对应的，该第二转换单元，具体用于：根据PQ电光转移函数，将该第二非线性亮度信号转换为该第二线性亮度信号。
本申请第三方面提供了一种视频信号处理的装置，该装置包括：处理器和存储器；该处理器用调用该存储器中的软件指令，以执行如下步骤：获得第一线性亮度信号，该第一线性亮度信号基于待处理视频信号对应的第一线性红绿蓝RGB信号获得；将该第一线性亮度信号转换为第一非线性亮度信号；对该第一非线性亮度信号进行分段亮度映射，以得到第二非线性亮度信号；将该第二非线性亮度信号转换为第二线性亮度信号；计算该第二线性亮度信号与该第一线性亮度信号的亮度增益；基于该亮度增益和该第一线性RGB信号的乘积，获得该待处理视频信号对应的RGB显示信号。
在一种可能的实施方式中，该待处理视频信号为感知量化PQ信号，该处理器，具体用于：对该PQ信号进行色彩空间转换，以获得第一非线性RGB信号；根据PQ电光转移函数，将该第一非线性RGB信号转换为该第一线性RGB信号；基于该第一线性RGB信号的各基色信号进行计算，以获得该第一线性亮度信号。
在一种可能的实施方式中，该待处理视频信号为混合对数伽玛HLG信号，该处理器具体用于：对该HLG信号进行色彩空间转换，以获得第二非线性RGB信号；根据HLG光电转移逆函数，将该第二非线性RGB信号转换为第二线性RGB信号；基于该第二线性RGB信号的各基色信号进行计算，以获得第三线性亮度信号；对该第三线性亮度信号进行亮度信号类型转换，以获得该第一线性亮度信号。
在一种可能的实施方式中，该处理器还用于：对该RGB显示信号进行色彩空间转换，以获得目标显示信号，其中，该目标显示信号的色彩格式和显示设备对应的色彩格式相同。
在一种可能的实施方式中，该处理器用于：对该RGB显示信号的各基色值叠加黑位 电平提升BlackLevelLift，以获得处理后的RGB显示信号，该BlackLevelLift为该显示设备的显示亮度的最小值；对应的，该处理器具体用于：对该处理后的RGB显示信号进行色彩空间转换。
在一种可能的实施方式中，该处理器具体用于：确定第一阈值和第二阈值，该第一阈值小于该第二阈值；当该第一非线性亮度信号的亮度值小于或等于该第一阈值时，该第二非线性亮度信号的亮度值等于该第一非线性亮度信号的亮度值；当该第一非线性亮度信号的亮度值大于该第一阈值，且小于或等于该第二阈值时，该第二非线性亮度信号的亮度值基于以该第一非线性亮度信号的亮度值为自变量的拟合曲线获得；当该第一非线性亮度信号的亮度值大于该第二阈值时，该第二非线性亮度信号的亮度值等于该显示设备对应的最大非线性显示亮度值。
在一种可能的实施方式中，该拟合曲线通过对该第一阈值和该第二阈值进行埃尔米特Hermite插值得到。
在一种可能的实施方式中，该拟合曲线存储在该存储器中。
在一种可能的实施方式中，该处理器具体用于：使用以下分段函数对该第一非线性亮度信号进行该亮度映射：
Figure PCTCN2019090566-appb-000005
其中，hmt(x)＝y 0×α 0(x)+y 1×α 1(x)+y′ 0β 0(x)+y′ 1β 1(x)
Figure PCTCN2019090566-appb-000006
e为该第一非线性亮度信号，f tm(e)为该第二非线性亮度信号，KP1为该第一阈值，KP2为该第二阈值，maxDL为该显示设备的最大非线性显示亮度值，maxSL为最大非线性源亮度值，x 0＝KP1，x 1＝maxSL，y 0＝KP1，y 1＝maxDL，y 0′＝1，y 1′＝0。
在一种可能的实施方式中，该分段函数存储在该存储器中。
在一种可能的实施方式中，该处理器具体用于：根据该第一非线性亮度信号的显示亮度范围和该显示设备的显示亮度范围的关系确定该第一阈值；将该第一非线性亮度信号的最大亮度值作为该第二阈值。
在一种可能的实施方式中，该处理器具体用于：基于预设的该第一非线性亮度信号和该第二非线性亮度信号的亮度值的映射关系，确定与该第一非线性亮度信号的亮度值相对应的第二非线性亮度信号的亮度值。该映射关系存储在该存储器中。
在一种可能的实施方式中，该处理器具体用于：根据PQ电光转移逆函数，将该第一线性亮度信号转换为该第一非线性亮度信号；对应的，该处理器，具体用于：根据PQ电光转移函数，将该第二非线性亮度信号转换为该第二线性亮度信号。
本申请第四方面提供了一种计算机可读存储介质，该计算机可读存储介质中存储有指令，当其在计算机或处理器上运行时，使得所述计算机或处理器执行如上述第一方面或者其任一种可能的实施方式中所述的方法。
本申请第五方面提供了一种包含指令的计算机程序产品，当其在计算机或处理器上运行时，使得所述计算机或处理器执行如上述第一方面或者其任一种可能的实施方式中所述的方法。
应理解，本申请实施例的第二至五方面以及各可能的实施方式与本申请实施例的第一方面以及各可能的实施方式，采用的技术手段相似，解决的技术问题相同，所能取得的技术效果可以参考对本申请实施例第一方面的相关描述，不再赘述。
附图说明
图1为本申请实施例提供的一种示例性的应用场景示意图；
图2为本申请实施例提供的一种示例性的应用场景的架构示意图；
图3为本申请实施例提供的另一种示例性的应用场景的架构示意图；
图4为本申请实施例提供的一种示例性的播放设备和显示设备的硬件架构示意图；
图5为本申请实施例提供的一种示例性的视频信号处理的方法流程图；
图6为本申请实施例提供的一种示例性的HDR视频信号亮度处理的方法流程图；
图7为本申请实施例提供的一种示例性的亮度映射曲线的示意图；
图8(a)为本申请实施例提供的一种示例性的静态元数据HDR tone mapping示意图；
图8(b)为本申请实施例提供的一种示例性的动态元数据HDR tone mapping示意图；
图9为本申请实施例提供的一种示例性的获得亮度映射曲线的方法流程图；
图10为本申请实施例提供的另一种示例性的HDR信号亮度处理的方法流程图；
图11为本申请实施例提供的一种示例性的PQ EOTF曲线(图左)和一种示例性的PQ EOTF -1曲线(图右)；
图12为本申请实施例提供的一种示例性的HLG OETF曲线(图左)和一种示例性的HLG OETF -1曲线(图右)；
图13为本申请实施例提供的一种示例性的视频信号处理的装置；
图14为本申请实施例提供的一种示例性的HDR终端技术方案处理流程图；
图15为本申请实施例提供的一种示例性的色域转换过程的示意图；
图16为本申请实施例提供的一种示例性的信号转换过程的示意图；
图17为本申请实施例提供的一种示例性的测试组网的示意图；
图18为本申请实施例提供的另一种示例性的测试组网的示意图；
图19为本申请实施例提供的一种示例性的HDR端到端系统架构图；
图20为本申请实施例提供的一种示例性的1000cd/m2的HLG曲线HDR视频端到端系统示意图；
图21为本申请实施例提供的一种示例性的非1000cd/m2的HLG曲线HDR视频端到端系统。
具体实施方式
本申请的说明书实施例和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元。方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
应当理解，在本申请中，“至少一个(项)”是指一个或者多个，“多个”是指两个或两个以上。“和/或”，用于描述关联对象的关联关系，表示可以存在三种关系，例如，“A和/或B”可以表示：只存在A，只存在B以及同时存在A和B三种情况，其中A，B可以是单数或者复数。字符“/”一般表示前后关联对象是一种“或”的关系。“以下至少一项(个)”或其类似表达，是指这些项中的任意组合，包括单项(个)或复数项(个)的任意组合。例如，a，b或c中的至少一项(个)，可以表示：a，b，c，“a和b”，“a和c”，“b和c”，或“a和b和c”，其中a，b，c可以是单个，也可以是多个。
首先，为了便于理解本申请实施例，对本申请实施例涉及的一些概念或术语进行解释。
基色值(color value)：一个对应于特定图像颜色分量(例如R、G、B或Y)的数值。
数字编码值(digital code value)：一个图像信号的数字表达值，数字编码值用于表示非线性基色值。
线性基色值(linear color value)：线性的基色值，与光强度成正比，在一种可选的情况中，其值应归一化到[0,1]，简称E。
非线性基色值(nonlinear color value)：非线性的基色值，是图像信息的归一化数字表达值，与数字编码值成正比，在一种可选的情况中，其值应归一化到[0,1]，简称E′。
电光转移函数(electro-optical transfer function，EOTF)：一种从非线性基色值到线性基色值之间的转换关系。
光电转移函数(optical-electro transfer function，OETF)：一种从线性基色值到非线性基色值之间的转换关系。
元数据(Metadata)：视频信号中携带的描述视频源信息的数据。
动态元数据(dynamic metadata)：与每帧图像相关联的元数据，该元数据随画面不同而改变。
静态元数据(static metadata)：与图像序列相关联的元数据，该元数据在图像序列内保持不变。
亮度信号(luma)：表示非线性基色信号的组合，符号为Y'。
亮度映射(luminance mapping)：源图像的亮度映射到目标系统的亮度。
色度体积(colour volume)：显示器在色度空间中所能呈现的色度和亮度形成的体积。
显示适配(display adaptation)：对视频信号进行处理，以适配目标显示器的显示特性。
源图像(source picture)：HDR前处理阶段输入的图像。
主控监视器(Mastering Display)：视频信号在编辑制作时使用的参考显示器，用来确定视频编辑制作的效果；
线性场景光(Linear Scene Light)信号：HDR视频技术中以内容为场景光的HDR视频信号，指的是相机/摄像头传感器捕获到的场景光，一般为相对值；线性场景光信号经过HLG编码之后得到HLG信号，HLG信号为一种场景光信号，HLG信号是非线性的；场景 光信号一般需要经过OOTF转换成显示光信号在显示设备上显示；
线性显示光(Linear Display Light)信号：HDR视频技术中以内容为显示光的HDR视频信号，指的是显示设备上发出的显示光，一般为绝对值，单位nits；线性显示光信号经过PQ编码之后得到PQ信号，PQ信号为一种显示光信号，PQ信号为非线性信号；显示光信号一般标准按照其绝对值亮度在显示设备上显示；
光-光转换曲线(OOTF)：视频技术中将一种光信号转换成另外一种光信号的曲线；
动态范围(Dynamic Range)：视频信号中最大亮度与最小亮度的比值；
亮度-色度-色度(Luma-Chroma-Chroma)：亮色分离视频信号的三个分量；
感知量化(Perceptual Quantizer，PQ)：一种HDR标准，也是一种HDR转换方程，PQ根据人的视觉能力决定。显示设备显示的视频信号通常为PQ编码格式的视频信号。
PQ EOTF曲线：将PQ编码的电信号转换成线性光信号，单位nits；转换公式为：
Figure PCTCN2019090566-appb-000007
其中，E′为输入电信号，取值范围[0,1]；固定参数值如下：
m1＝2610/16384＝0.1593017578125；
m2＝2523/4096x128＝78.84375；
c1＝3424/4096＝0.8359375＝c3-c2+1；
c2＝2413/4096x32＝18.8515625；
c3＝2392/4096x32＝18.6875；
PQ EOTF曲线如图11左图所示：输入是[0,1]范围的电信号，输出是[0,10000]nits的线性光信号；
PQ EOTF -1曲线：PQ EOTF的逆曲线；物理含义是将[0,10000]nits的线性光信号转换成PQ编码的电信号；转换公式为：
Figure PCTCN2019090566-appb-000008
PQ EOTF -1曲线如图11右图所示：输入是[0,10000]nits的线性光信号，输出是[0,1]范围的电信号；
色域(Color Gamut)：某个色彩空间包含颜色的范围，相关色域标准有BT.709,BT.2020.
混合对数伽马(Hybrid Log Gamma，HLG)：一种HDR标准，照相机、摄影机、图像传感器或其他种类的图像采集设备采集到的视频信号是HLG编码格式的视频信号。
HLG OETF曲线：对线性场景光信号进行HLG编码转换成非线性电信号的曲线，转换公式如下所示：
Figure PCTCN2019090566-appb-000009
其中E是输入线性场景光信号，范围[0，1]；E′是输出非线性电信号，范围[0,1]；
固定参数a＝0.17883277,b＝0.28466892,c＝0.55991073，如图12左所示，为HLG OETF曲线的一种示例图。
HLG OETF -1曲线：HLG OETF的逆曲线，将HLG编码的非线性电信号的转换成线性场景光信号，示例性的，转换公式如下所示：
Figure PCTCN2019090566-appb-000010
如图12右图所示，为HLG OETF -1曲线的一种示例图，其中E′是输入的非线性电信号，范围[0,1]；E是输出的线性场景光信号，范围[0，1]。
线性空间：本申请中线性空间指的是线性光信号所在的空间；
非线性空间：本申请中非线性空间指的是将线性光信号利用非线性曲线进行转换后所在的空间；HDR常用的非线性曲线有PQ EOTF-1曲线、HLG OETF曲线等，SDR常用的非线性曲线有伽马曲线；一般认为线性光信号经过上述非线性曲线编码之后，相对于人眼是视觉线性的。应当理解，非线性空间可以认为是视觉线性空间。
伽马校正(Gamma Correction)：伽马校正是对图像进行非线性色调编辑的方法，可以检出图像信号中的深色部分和浅色部分，并使两者比例增大，从而提高图像对比度效果。目前的显示屏、摄影胶片和许多电子照相机的光电转换特性均可以是非线性的。这些非线性部件的输出与输入之间的关系可以用一个幂函数来表示，即：输出＝(输入) γ。
设备输出的色彩值进行非线性转换是由于人类的视觉系统不是线性的，人类是通过比较来感知视觉刺激。外界以一定的比例加强刺激，对人来说，这个刺激才是均匀增长的。因此，对人类的感知来说，以等比数列增加的物理量，是均匀的。为了将输入的颜色按照人类视觉规律进行显示，需要经过上述幂函数形式的非线性转换，将线性的色彩值转换为非线性的色彩值。gamma的取值γ可以是根据色彩空间的光电转换曲线确定的。
色彩空间(Color Space)：颜色可以是眼睛对于不同频率的光线的不同感受，也可以表示客观存在的不同频率的光。色彩空间是人们建立起用来表示色彩的坐标系统所定义的色彩范围。色域与色彩模型一起定义一个色彩空间。其中，色彩模型是用一组颜色成分表示颜色的抽象数学模型。色彩模型例如可以包括三原色光模式(red green blue，RGB)、印刷四色模式(cyan magenta yellow key plate，CMYK)。色域是指一个系统能够产生的颜色的总合。示例性的，Adobe RGB和sRGB是两个基于RGB模型的不同的色彩空间。
每台设备例如显示器或打印机都有自己的色彩空间，且只能生成其色域内的颜色。将图像从一台设备移至另一台设备时，由于每台设备按照自己的色彩空间转换并显示RGB或CMYK，图像在不同的设备上的颜色可能会发生变化。
下面介绍几种常用的色彩空间。
sRGB色彩空间(standard Red Green Blue color space)：是惠普与微软于1996年一起开发的用于显示器、打印机以及因特网的一种标准RGB色彩空间。它提供一种标准方法来定义色彩，让显示、打印和扫描等各种计算机外部设备与应用软件对于色彩有一个共通的语言。sRGB的色彩空间是基于独立的色彩坐标，可以使色彩在不同的设备使用传输中对应于同一色彩坐标体系，而不受这些设备各自具有的不同色彩坐标的影响。但是sRGB的色域空间比较小。sRGB定义了红色、绿色与蓝色三原色的颜色，其中，三原色中一个颜色的色彩值取最大值，且其它两个颜色的色彩值都为零时所对应的颜色表示该一个颜色。示例性的，红色、绿色与蓝色三原色中，色彩值R、G和B的取值均为0-255，则当R、G取值均为零时，B取值为255时所对应的颜色表示蓝色。
YCC色彩空间：本申请中表示亮色分离的颜色空间，YCC三个分量分别代表亮度-色度-色度(Luma-Chroma-Chroma)；常见的YCC空间视频信号有YUV、YCbCr、ICtCp等；
应当理解，不同的色彩空间之间可以进行转换。
保留位(reserved_bits)：位流中的“保留位”表明保留了一些语法单元用于将来对本部分的扩展，解码处理应忽略这些位。“保留位”不应出现从任意字节对齐位置开始的21个以上连续的‘0’。
标记位(marker_bit)：指该位的值应为‘1’。
HDR视频可显示的亮度较大，在一种可选的情况中，SDR视频信号最大亮度100nits，HDR视频信号最大亮度1000nits以上，然而大量现有显示设备能够显示的亮度范围达不到HDR视频的亮度；因此在显示HDR视频信号时，需要根据显示设备的显示能力对HDR视频信号亮度进行处理，使HDR视频信号匹配显示设备可显示的亮度范围，适合在当前设备上显示；ITU BT.2100中定义的PQ信号和HLG信号是国际公认的两种HDR信号源，被多个国家地区纳入标准。
如图1所示，为本申请实施例提供的一种示例性的应用场景示意图。在该应用场景中，播放设备102完成视频流101的接收和解码，播放设备102通过高清晰度多媒体接口(High Definition Multimedia Interface，HDMI)将解码得到的视频或音频数据发送给显示设备103进行显示或播放，从而使用户可以欣赏到视频或音频内容。
示例性的，该视频流101可以来自网站流媒体、远端网络设备、互联网、光纤网络等，该视频流可以是动态元数据HDR视频数据，也可以是静态元数据HDR视频数据；示例性的，视频流101可以是传输流(Transport Stream，TS)格式的数据流，TS中可以包含视频流、音频流、字幕数据包等，视频流也可以采用其他类似格式的数据流，例如流媒体也可以使用(Matroska Video File，MKV)的格式来同时封装音频数据、视频数据和字幕数据等，类似的还有音频视频交织(Audio Video Interleaved，AVI)格式，本申请对音视频流的传输格式不做限定；示例性的，视频流中可以包括：HDR视频数据，以及用于描述HDR视频的元数据。在这种情况下，元数据和HDR视频数据均压缩在视频流中。在另一种可选的情况中，TS中可以包含视频流、音频流、字幕数据包、用于描述HDR视频的元数据，此种情况下，用于描述HDR视频数据的元数据放在TS中，而没有压缩在视频流中。元数据包含了对视频图像数据的描述。示例性的，静态元数据描述的是整条视频的制作环境，可以包括：视频制作调色和校正所用的监视器的相关信息、峰值亮度、黑位、RGB三色坐标和白色点坐标等。示例性的，动态元数据中一般包含了对视频图像每一帧内容的描述，例如可以包含图像的最高亮度、最低亮度、平均亮度等；可选的，动态元数据还可以包括某一帧图像与显示屏的参考映射曲线，应当理解，动态元数据中包含的参考映射曲线随着视频图像的变化而有所不同。
在一种可选的方案中，播放设备102可以是机顶盒(Set Top Box，STB)、多媒体播放器等，STB主要包含了较多的安全性功能，例如插卡计费、视频的加密解密等，一些质量较高的视频存在数字权利保护机制，需要STB侧解密之后，TV侧才能观看；STB侧接收到的视频数据通常是经过编码的视频数据，STB还具有解码功能，STB通常会将视频数据解码之后再发送给TV侧显示。如图3所示，为本申请实施例提供的一种由播放设备完成对视频数据的处理，然后送给显示设备显示的一种应用场景的架构示意图，STB解析接收到的TS，得到视频数据、音频数据、元数据metadata等，示例性的，STB接收的视频数据可以是HDR HLG视频信号、HDR PQ视频信号或者SDR视频信号。然后，STB对视频 数据进行解码、显示亮度处理、颜色饱和度处理、色域处理等，使得HDR视频信号适配显示屏的亮度范围，然后将处理之后的视频数据通过有线或无线的HDMI、Display Port传输等方式传送给显示设备103，可选的，经过播放设备处理之后的视频数据与显示设备能够显示的亮度范围相匹配，处理后的视频数据的类型与显示设备的种类相关，可以是HDR视频数据，也可以是SDR视频数据。如图3所示，计算亮度映射曲线由主处理器实现，一般利用软件完成的；提前计算好的亮度映射曲线存储在存储器中，视频处理器中的亮度处理单元调用存储器中的亮度映射曲线，完成对视频画面的处理，可选的，视频处理器可以为专用的集成硬件或专用电路也可以为跑在专用芯片上的多个软件模块；应当理解，图3只是一种示例，对播放设备的架构并不构成限定，在实际使用中，各功能模块的实现可按照实际应用场景调整，既可以全部利用视频处理器实现、也可以全部利用主处理器软件实现，还可以利用其它专用芯片如DSP、FPGA等硬件结合软件方式实现。
在一种可选的情况中，包含视频数据的视频流等可以直接传送给显示设备103如TV等，然后在显示设备内部对视频数据进行视频解码、亮度映射处理、颜色饱和度处理、色域处理等，如图2所示为本申请实施例提供的一种由显示设备完成对视频数据的处理之后进行显示的应用场景架构示意图。在一种可选的情况中，计算亮度映射曲线由主处理器实现，一般利用软件指令完成；将提前计算好的亮度映射曲线存储在存储器中，视频处理器中的亮度处理单元调用存储器中的亮度映射曲线，完成对视频画面的处理；应当理解，在实际使用中，各功能模块的实现可按照实际应用场景调整，既可以全部利用视频处理器实现、也可以全部利用主处理器软件实现，还可以利用其它专用芯片如DSP、FPGA等硬件结合软件方式实现。
示例性的，显示设备103可以是TV、电脑显示器或者任何具有显示功能的固定终端或移动终端。TV对STB发送来的视频数据进行显示处理后显示在屏幕上。示例性的，显示处理可以是TV根据预设的亮度映射关系对视频的亮度进行调整，使得视频内容与TV的显示能力相适应。在一种可选的情况中，当STB传来的视频为具有动态元数据的HDR视频，TV为具有HDR显示功能的TV，TV基于动态元数据中包含的多组亮度映射关系对不同的视频画面进行亮度映射tone mapping处理，使各种不同画面都能够以最佳方式显示和呈现。可选的，TV还可以为SDR TV或HDR PQ TV。
下面从处理器硬件的角度对本申请实施例中的一种播放设备和显示设备分别进行描述。
如图4所示，为本申请实施例提供的一种播放设备和显示设备的硬件架构示意图。
可选的，播放设备102包括至少一个中央处理器CPU、至少一个存储器、GPU、解码器、专用的视频/图形处理器、输入接口和HDMI发送器等。可选的，播放设备还可以包括微处理器和微控制器(Microcontroller Unit，MCU)等。在一种可选的情况中，播放设备的上述各个部分通过连接器相耦合，该连接器可包括各类接口、传输线或总线等，本实施例对此不做限定。所述连接器可包括各类接口、传输线或总线等，本实施例对此不做限定。在一种可选的情况中，上述各部分集成在同一个芯片上，共同构成播放设备的核心芯片；在另一种可选的情况中，CPU、GPU、解码器、输入接口以及HDMI发送器集成在一个芯片上，该芯片内部的各部分通过总线访问外部的存储器。专用视频/图形处理器可以与CPU集成在同一个芯片上，也可以作为单独的处理器芯片存在。在本申请实施例中涉及的芯片 是以集成电路工艺制造在同一个半导体衬底上的系统，也叫半导体芯片，其可以是利用集成电路工艺制作在所述衬底(通常是例如硅一类的半导体材料)上形成的集成电路的集合，其外层通常被半导体封装材料封装。所述集成电路可以包括各类功能器件，每一类功能器件包括逻辑门电路、金属氧化物半导体(Metal-Oxide-Semiconductor，MOS)晶体管、双极晶体管或二极管等晶体管，也可包括电容、电阻或电感等其他部件。每个功能器件可以独立工作或者在必要的驱动软件的作用下工作，可以实现通信、运算、或存储等各类功能。
示例性的，CPU可以用于实现本申请实施例的部分或全部运算，例如可以实现图像的tone mapping、视频数据的解复用、解码、亮度映射曲线tone-mapping curve的计算、颜色饱和度曲线的计算等；可选的，CPU可以是一个单核(single-CPU)处理器或多核(multi-CPU)处理器；可选的，CPU可以是多个处理器构成的处理器组，多个处理器之间通过一个或多个总线彼此耦合。在一种可选的情况中，对于图像和视频的处理部分由GPU完成，部分由专用视频/图形处理器完成，还有可能是由跑在通用CPU或GPU上的软件代码完成。
存储器，可用于存储计算机程序指令，包括操作系统(Operation System，OS)、各种用户应用程序、以及用于执行本申请方案的程序代码在内的各类计算机程序代码，例如可以用于存储亮度映射曲线或者亮度映射查找表LUT；存储器还可以用于存储视频数据、音频数据和字幕数据等；CPU可以用于执行存储器中存储的计算机程序代码，以实现本申请实施例中的方法，被执行的各类程序代码可以认为是CPU的驱动程序。可选的，存储器302可以是非掉电易失性存储器，例如是嵌入式多媒体卡(Embedded Multi Media Card，EMMC)、通用闪存存储(Universal Flash Storage，UFS)或只读存储器(Read-Only Memory，ROM)，或者是可存储静态信息和指令的其他类型的静态存储设备，还可以是掉电易失性存储器(volatile memory)，例如随机存取存储器(Random Access Memory，RAM)或者可存储信息和指令的其他类型的动态存储设备，也可以是电可擦可编程只读存储器(Electrically Erasable Programmable Read-Only Memory，EEPROM)、只读光盘(Compact Disc Read-Only Memory，CD-ROM)或其他光盘存储、光碟存储(包括压缩光碟、激光碟、光碟、数字通用光碟、蓝光光碟等)、磁盘存储介质或者其他磁存储设备、或者能够用于携带或存储具有指令或数据结构形式的程序代码并能够由计算机存取的任何其他计算机可读存储介质，但不限于此。
输入接口用于接收传输流，播放设备的输入接口可以是网络接口，例如可以是WIFI或者以太网接口，该输入接口也可以是广播电视的端子例如tunner，该输入接口还可以是通用串行总线(Universal Serial Bus，USB)接口。
在一种可选的情况中，CPU执行相关的代码对传输流进行解复用，得到视频数据和字幕数据等，解码器对视频数据流进行解码得到视频数据和metadata，视频/图形处理器完成对视频数据的亮度映射处理、颜色饱和度处理、色彩空间转换、色预处理、场景光信号和显示光信号转换、线性空间和非线性空间转换等，可选的，还可以完成对亮度映射曲线、饱和度映射曲线的计算。HDMI发送器对解码后的视频数据、metadata和字幕数据分别进行封装，并将封装之后的数据包/信息帧通过HDMI数据通道传送给显示设备103。
可选的，显示设备103包括HDMI接收器、至少一个中央处理器CPU、至少一个存储器、GPU、解码器、专用的视频/图形处理器以及V By One接口等。该显示设备还包括显示屏(图中未示出)，可选的，V By One接口耦合至显示屏，V By One接口是一种面向图 像传输开发的数字接口标准。可选的，图4中所示的显示设备103可以是一个集成的显示器芯片，接收的视频数据在该显示器芯片上处理之后发送给显示屏进行显示。
对于CPU和存储器的具体情况请参考对播放设备侧的CPU和存储器的描述，此处不再赘述。HDMI发送器将视频数据帧、元数据信息帧和字幕信息帧以及其他的信息帧或数据包分离传输给显示设备。应当理解，HDMI内部有多条通道，其中一部分通道用于传输数据信息，还有一部分用于传输时钟、校验信号、电源信号以及地信号等控制信息。在一种可选的情况中，数据通道是分时复用的，各种数据包不能同时传送。通道单位时间内传输的数据量受工作频率的限制，通道单位时间内能够传输的最大数据量为HDMI通道的带宽，示例性的，HDMI2.1的带宽是18G bps(bit per second)。可选的，传输接口分时传输HDR视频数据帧、元数据信息帧和字幕信息帧。可选的，传输接口对应多个传输频段，传输接口分频传输视频数据帧、元数据帧和字幕信息帧；可选的，传输接口对应多个传输通道，传输接口分通道传输视频数据帧、元数据帧和字幕信息帧。对视频数据的tone mapping处理可以由GPU完成，也可以由专用的视频/图形处理器完成。亮度映射可以由专用的视频/图形处理器完成、也可以由跑在CPU或GPU上的软件代码完成。在一种可选的情况中，视频/图像处理器将显示亮度处理之后的视频数据，通过V By One接口传送给显示屏进行显示。
示例性的，显示屏可以是液晶显示器(Liquid Crystal Display，LCD)，发光二级管(Light Emitting Diode，LED)显示器，有机发光二极管(Organic Light-Emitting Diode，OLED)显示屏、阴极射线管(Cathode Ray Tube，CRT)显示屏等。
如图5所示，为本申请实施例提供的一种视频信号处理的方法流程图。应当理解，为了便于描述，图5以步骤的形式对方法进行描述，虽然在方法流程图5中示出了方法的顺序，但是在某些情况下，可以以不同于此处的顺序执行所描述的步骤。
该视频信号处理的方法包括：
501、获得第一线性亮度信号；
该第一线性亮度信号基于待处理视频信号对应的第一线性RGB信号获得。示例性的，该第一线性RGB信号为线性显示光信号，该第一线性亮度信号为线性显示光亮度信号，应当理解，亮度信号为待处理视频信号的亮度分量，在一种可选的情况中，第一线性亮度信号基于第一线性RGB信号的三个基色信号R、G、B计算获得。
示例性的，待处理的视频信号可以为PQ信号，该PQ信号可以为YUV空间的PQ信号，在获得线性亮度信号之前需要将PQ信号从YUV空间转换到RGB空间，得到第一非线性RGB信号；可选的，根据PQ电光转移函数，将第一非线性RGB信号转换为第一线性RGB信号，然后基于第一线性RGB信号的各基色信号进行计算，获得第一线性亮度信号，在该情况中，第一线性亮度信号为显示光亮度信号。
示例性的，待处理的视频信号可以为HLG信号，该HLG信号可以为YUV空间的HLG信号，在获得线性亮度信号之前需要将HLG信号从YUV空间转换到RGB空间，得到第二非线性RGB信号；可选的，根据HLG电光转移逆函数，将第二非线性RGB信号转换为第二线性RGB信号，该第二线性RGB信号为线性场景光信号，然后基于第二线性RGB信号的各基色信号进行计算，获得第三线性亮度信号，应当理解，由于HLG信号为场景光信号，获得的第三线性亮度信号为场景光亮度信号，需要将第三线性亮度信号进行亮度 信号类型转换，得到第一线性亮度信号。示例性的，第一线性亮度信号为显示光亮度信号。
502、将第一线性亮度信号转换为第一非线性亮度信号；
本申请实施例在非线性空间中进行亮度映射，或者说，本申请实施例中，亮度映射的输入为线性的亮度信号，因此需要将线性亮度信号转换为非线性亮度信号。
示例性的，可以根据PQ电光转移逆函数，将第一线性亮度信号转换为第一非线性亮度信号。也可以采用其他转换函数或转换曲线完成第一线性亮度信号和第一非线性亮度信号的转换。第一非线性亮度信号为亮度映射前的非线性亮度信号。
503、对所述第一非线性亮度信号进行分段亮度映射，得到第二非线性亮度信号；
可选的，可以基于预设的亮度映射查找表对第一非线性亮度信号进行分段亮度映射，该亮度映射查找表可以存储在存储器中，该亮度映射查找表中为若干组离散的坐标点，输入一个第一非线性亮度信号的亮度值对应输出一个第二非线性亮度信号的亮度值，可选的，该亮度映射查找表中的映射数据可以通过预先的实验获得；
可选的，可以基于分段函数对第一非线性亮度信号进行分段亮度映射，在一种可选的情况中，可以将第一非线性亮度信号根据亮度值分成三段，每段的亮度映射函数均不同；示例性的，确定第一阈值和第二阈值，第一阈值小于第二阈值；当第一非线性亮度信号的亮度值小于或等于第一阈值时，第二非线性亮度信号的亮度值等于第一非线性亮度信号的亮度值；当第一非线性亮度信号的亮度值大于第一阈值，且小于或等于第二阈值时，第二非线性亮度信号的亮度值基于以第一非线性亮度信号的亮度值为自变量的拟合曲线获得；当第一非线性亮度信号的亮度值大于第二阈值时，所述非线性亮度信号的亮度值等于显示设备对应的最大非线性显示亮度值。在一种可选的情况中，拟合曲线通过对第一阈值和第二阈值进行Hermite插值得到。
示例性的，以下为本申请实施例提供的几种分段亮度映射曲线。
情况1：亮度映射前的亮度范围为0-1000nits，亮度映射后的亮度范围为0-100nits：
则分段亮度映射曲线可表示为如下分段函数：
Figure PCTCN2019090566-appb-000011
其中函数hmt()定义如下：
hmt(x)＝0.2643×α 0(x)+0.5081×α 1(x)+β 0(x)   (6)
Figure PCTCN2019090566-appb-000012
情况2：亮度映射前的亮度范围为0-1000nits，亮度映射后的亮度范围为0-150nits：
则分段亮度映射曲线可表示为如下分段函数：
Figure PCTCN2019090566-appb-000013
其中函数hmt()定义如下：
hmt(x)＝0.3468×α 0(x)+0.5493×α 1(x)+β 0(x)    (9)
Figure PCTCN2019090566-appb-000014
情况3：亮度映射前的亮度范围为0-1000nits，亮度映射后的亮度范围为0-200nits：
则分段亮度映射曲线可表示为如下分段函数：
Figure PCTCN2019090566-appb-000015
其中函数hmt()定义如下：
hmt(x)＝0.4064×α 0(x)+0.5791×α 1(x)+β 0(x)    (12)
Figure PCTCN2019090566-appb-000016
情况4：亮度映射前的亮度范围为0-1000nits，亮度映射后的亮度范围为0-250nits：
则分段亮度映射曲线可表示为如下分段函数：
Figure PCTCN2019090566-appb-000017
其中函数hmt()定义如下：
hmt(x)＝0.4533×α 0(x)+06026×α 1(x)+β 0(x)    (15)
Figure PCTCN2019090566-appb-000018
情况5：亮度映射前的亮度范围为0-1000nits，亮度映射后的亮度范围为0-300nits：
则分段亮度映射曲线可表示为如下分段函数：
Figure PCTCN2019090566-appb-000019
其中函数hmt()定义如下：
hmt(x)＝0.4919×α 0(x)+0.6219×α 1(x)+β 0(x)    (18)
Figure PCTCN2019090566-appb-000020
情况6：亮度映射前的亮度范围为0-1000nits，亮度映射后的亮度范围为0-350nits：
则分段亮度映射曲线可表示为如下分段函数：
Figure PCTCN2019090566-appb-000021
其中函数hmt()定义如下：
hmt(x)＝0.5247×α 0(x)+0.6383×α 1(x)+β 0(x)    (21)
Figure PCTCN2019090566-appb-000022
情况7：亮度映射前的亮度范围为0-1000nits，亮度映射后的亮度范围为0-400nits：
则分段亮度映射曲线可表示为如下分段函数：
Figure PCTCN2019090566-appb-000023
其中函数hmt()定义如下：
hmt(x)＝0.5533×α 0(x)+0.6526×α 1(x)+β 0(x)    (24)
Figure PCTCN2019090566-appb-000024
可选的，可以基于亮度映射曲线对第一非线性亮度信号进行分段亮度映射，该亮度映射曲线可以是分段曲线，应当理解，分段亮度映射曲线可以认为是分段函数的图示，查找表中离散数据可以为亮度映射曲线上的坐标点。
504、将第二非线性亮度信号转换为第二线性亮度信号；
示例性的，可以根据PQ电光转移函数，将第二非线性亮度信号转换为第二线性亮度信号。也可以采用其他转换函数或转换曲线完成第二非线性亮度信号和第二线性亮度信号的转换。第二非线性亮度信号为亮度映射后的非线性亮度信号。
505、计算第二线性亮度信号与第一线性亮度信号的亮度增益；
示例性的，该亮度增益为第二线性亮度信号与第一线性亮度信号的比值。
506、基于亮度增益和第一线性RGB信号的乘积，获得待处理视频信号对应的RGB 显示信号。
具体的，该亮度增益与第一线性RGB信号的三个基色分量R、G、B分别相乘，得到RGB显示信号，可选的，该RGB显示信号可用于显示设备的显示。
在一种可选的情况中，显示设备可显示的色彩格式与RGB不同，则该方法还包括：
对该RGB显示信号进行色彩空间转换，得到目标显示信号，目标显示信号的色彩格式和显示设备对应的色彩格式相同。
在一种可能的实施方式中，在获得RGB显示信号之后，在RGB显示信号的每个基色分量R、G、B上分别叠加黑位电平抬升BlackLevelLift，BlackLevelLift为显示设备的显示亮度的最小值。进一步的，对处理后的显示RGB信号进行色彩空间转换，得到与显示设备对应的色彩格式相同的目标显示信号。
下面分别以输入为HDR PQ信号和输入为HDR HLG信号为例对本申请实施例提供的视频信号处理方法进行说明。
如图6所示，为本申请实施例提供的一种HDR视频数据亮度处理的方法流程示意图。
应当理解，线性场景光信号和线性显示光信号为视频信号的两种形态，线性场景光信号为camera或其他图像采集设备或视频采集设备捕获到的视频信号，线性显示光信号为显示设备显示的视频信号，亮度信号为视频信号中表征亮度的分量，基于线性场景光信号获得的是线性场景亮度信号，基于线性显示光信号获得的是线性显示光亮度信号。具体的，该方法包括：
600、计算亮度映射曲线TM_Curve。
亮度映射曲线可以是任意作用于所选非线性空间上的亮度映射曲线，示例性的，在本申请实施例中该非线性空间可以为PQ曲线空间。如图7所示为在非线性空间(PQ EOTF -1曲线)上生成的亮度映射曲线的一种示例，该亮度映射曲线横轴为亮度映射前的非线性编码亮度信号，纵轴为亮度映射后非线性编码亮度信号，示例性的，横坐标和纵坐标的取值范围均为[0,1]，代表将亮度范围为[0,10000]nits的线性光信号经过PQ曲线编码之后的电信号的范围为[0,1](可参考图11右图，将亮度范围0-10000nits的亮度信号经过PQ曲线编码后得到0-1的电信号)，也即，横坐标表示亮度映射前的亮度信号的亮度范围为[0,10000]nits，纵坐标表示亮度映射后的亮度信号的亮度范围，图7所示的曲线表示将亮度范围为[0,10000]nits的亮度信号映射到亮度范围[0,300]nits内。
可选的，该亮度映射曲线可以表示为一个亮度映射曲线公式，如y＝1-e -x；其中x表示输入的显示亮度，y表示输出的显示亮度；在一种可选的情况中，该亮度映射的曲线为一个分段函数，示例性的，亮度映射曲线的公式可以如下所示：
Figure PCTCN2019090566-appb-000025
其中函数hmt()定义如下：
hmt(x)＝0.2643×α 0(x)+0.5081×α 1(x)+β 0(x)    (27)
Figure PCTCN2019090566-appb-000026
应当理解，上述公式将原始的亮度范围0-1000nits映射到0-100nits的亮度范围内。
在一种可选的情况中，采用一维查找表LUT来表示该亮度映射曲线，一维查找表中存储有一定数量的离散的坐标点TM_Curve＝{TM_Curve_x,TM_Curve_y}，这些离散的坐标点为亮度映射曲线上的采样点，其中，TM_Curve_x＝(x0,x1,…xn)是n个采样点的横坐标，即亮度映射曲线的输入，代表亮度映射前的亮度信号；TM_Curve_y＝(y0,y1,…yn)是n个采样点的纵坐标，即亮度映射曲线的输出，代表亮度映射后的亮度信号。应当理解，由于查找表中的映射关系是一组离散的点，如果输入的亮度值没有在这些离散的点中，可以基于已知的亮度值插值得到输入的亮度值；在一种可选的情况中，位于一个范围内的输入的亮度值对应同一个输出的亮度值，例如x0-x3均对应y0，x4-x6均对应y1等。这些离散的坐标点可以是通过预先实验测量得到的，例如输入亮度值，在显示屏上测量得到输入亮度值对应的显示亮度值；在一种可选的情况中，可以得到输入的亮度值与显示屏显示的亮度值的对应关系的拟合函数，根据拟合函数计算得到查找表中的离散坐标点。
应当理解，亮度映射的实质是对视频图像的显示亮度范围进行调节，从一种显示亮度范围映射到另外一种显示亮度范围中，例如可以将视频图像的亮度范围调节到显示屏能够显示的亮度范围内，TM_Curve曲线的输入信号是亮度映射前的亮度信号，输出是亮度映射后的亮度信号，为了便于理解TM_Curve对视频图像的作用，下面分别给出一组包含在静态元数据HDR中的TM_Curve和一组动态元数据HDR中的TM_Curve的tone mapping示意图，分别如图8(a)和图8(b)所示，应当理解，tone mapping可以是亮度映射的一种实现方式，或者说tone mapping的过程在某种情况下可以等同于亮度映射的过程。
静态元数据HDR视频用固定的映射方法来处理所有的视频图像，如图8(a)所示，静态元数据HDR视频具有一个固定的映射曲线，该映射曲线输入的亮度范围为0-4000尼特nit，输出的亮度范围为0-500nit，TV对最高亮度为200nit、最高亮度为1500nit和最高亮度为4000nit的三帧图像，均采用该映射曲线进行tone mapping处理，处理之后显示在显示器上。由于映射关系曲线与前两帧图像实际具有的亮度范围不相符，前两帧图像丢失了大部分图像细节，整体偏暗，显示效果较差。动态元数据HDR视频具有多组映射曲线，在显示视频图像的过程中根据图像的实际亮度调整所使用的映射曲线，如图8(b)所示，给出了三组不同的映射曲线，映射曲线1输入的亮度范围为0-500尼特nit，输出的亮度范围为0-500nit，映射曲线2输入的亮度范围为0-1500尼特nit，输出的亮度范围为0-500nit，映射曲线3输入的亮度范围为0-4000尼特nit，输出的亮度范围为0-500nit，TV根据图像帧的实际亮度范围选择合适的映射曲线进行tone mapping处理，使得不同亮度的图像都能以最佳方式呈现在屏幕上，例如选取映射曲线2对最高亮度为1500nit的图像进行亮度映射，处理之后的图像很好的保留了图像细节。应当理解，上述映射曲线的输入亮度表示的是视频信号源所表示的亮度；而输出亮度则是显示设备如TV所能真正显示出的亮度。对于HDR视频来说，一般在专业影视制作环境下制作的视频信号所表示的亮度范围比消费类电视机所能显示的亮度范围要大。tone mapping为一种将输入的视频信号的亮度范围映 射和匹配到显示设备显示的亮度范围的技术。
601、计算线性显示光信号RdGdBd。
示例性的，本申请实施例中输入的视频信号为YUV空间的HDR PQ信号，应当理解，对线性显示光信号进行PQ编码得到PQ信号，PQ信号为一种显示光信号，且PQ信号是一种非线性信号。经过色彩空间转换，将输入的视频信号从YUV空间转换到RGB空间，得到RGB色彩空间中的非线性显示光信号R’dG’dB’d；进一步的，将非线性显示光信号R’dG’dB’d转换成线性显示光信号RdGdBd，在一种可选的情况中，基于PQ EOTF曲线将非线性显示光信号R’dG’dB’d转换成线性显示光信号RdGdBd。
应当理解，在这种情况中，认为YUV转换至RGB属于色彩空间转换，而非线性RGB转线性RGB信号并没有进行色彩空间的转换，即在这种情况中，非线性RGB信号和线性RGB信号同属RGB色彩空间。色彩空间转换采用标准的色彩空间转换过程；其中Rd、Gd、Bd＝[0,10000]；在一种可选的情况中，也可以处理其他YCC空间的HDR PQ信号。
应当理解，在一种可选的情况中，也可以认为将输入的视频信号从YUV信号转换为非线性显示光信号R’dG’dB’d为一次色彩空间转换，将非线性显示光信号R’dG’dB’d转换为线性显示光信号RdGdBd也可以认为是一次色彩空间转换，在这种情况中，非线性RGB显示光信号和线性RGB显示光信号属于不同的色彩空间。
602、计算线性显示光亮度Yd。
根据以下公式计算显示光亮度Yd，
Yd＝(cr*Rd+cg*Gd+cb*Bd)   (29)
Yd为线性的亮度信号，亮度信号为表征显示光信号的亮度的量，即也可以认为亮度信号为表征视频信号的亮度的分量；其中，参数cr，cg，cb的选择与线性显示光信号RdGdBd的色域有关，当显示信号RdGdBd的色域为BT.2020时，计算参数cr＝0.2627，cg＝0.6780，cb＝0.0593；当线性显示光信号的色域不同时，对应的计算参数也会不同，因此计算亮度时需要根据光信号所处的色域选取对应色域下线性亮度计算参数。
603、将显示光亮度Yd转换到非线性空间NL_Yd；
Yd为线性空间上的显示光亮度，示例性的，将显示光亮度Yd利用PQ EOTF -1曲线转换到非线性空间得到非线性显示光亮度，NL_Yd＝PQ_EOTF -1(Yd)，该过程相当于将将亮度范围0-10000nits的线性显示光亮度信号Yd经过PQ曲线编码后得到0-1的电信号，该电信号即为非线性空间的非线性显示光亮度NL_Yd；应当理解，非线性空间转换曲线还可以采用其他非线性转换曲线。
604、对非线性显示光亮度信号NL_Yd进行亮度映射；
利用预先存储的亮度映射曲线TM_Curve，对非线性显示光亮度信号NL_Yd进行映射，得到映射后非线性亮度NL_Yt；应当理解，这里的亮度映射曲线TM_Curve可以是步骤500中的TM_Curve；非线性显示光亮度信号NL_Yd为亮度映射前的亮度信号，也可以说是源图像信号的非线性显示光亮度信号，NL_Yt为亮度映射后的非线性显示光亮度信号，也可以说是与显示设备的显示能力相适应的亮度信号，或者说是目标亮度信号。示例性的，亮度映射曲线TM_Curve通过计算公式表征时，输入一个非线性显示光亮度信号NL_Yd，基于该亮度映射曲线的计算公式获得一个对应的亮度映射后的非线性亮度NL_Yt。本实施例选取一维查找表LUT来实现亮度映射曲线，一维查找表的具体情况可参考步骤500中对一 维查找表的说明；可选的，在实际映射中可以采取查表线性插值方法；也可采用其他插值方法；线性插值是一种针对一维查找表LUT的插值方法，它根据一维数据序列中需要插值的点的左右邻近两个数据点来进行数值的估计。
605、将映射后的非线性显示光亮度信号NL_Yt转换到线性空间，得到线性显示光亮度信号Yt；
在一种可选的情况中，利用PQ EOTF曲线将映射后的非线性亮度NL_Yt转换到线性空间得到映射后的线性亮度Yt，即Yt＝PQ_EOTF(NL_Yt)；即将范围取[0,1]的非线性的电信号转换到范围为[0,10000]nits的线性显示光亮度信号。应当理解，可以采用PQ EOTF曲线以外的曲线将NL_Yt转换得到Yt。
606、计算线性亮度增益K；
示例性的，
K＝Yt/Yd   (30)
即K表征在亮度映射后的线性显示光亮度信号Yt和亮度映射前的线性显示光亮度信号Yd的比值，可用于衡量亮度映射前后显示光亮度的变动；
607、计算处理后的线性显示光信号；
(Rt,Gt,Bt)＝K*(Rd,Gd,Bd)+(BLoffset,BLoffset,BLoffset)    (31)
Rt、Gt、Bt分别为亮度映射处理后的线性显示光信号的红色分量、绿色分量和蓝色分量；BLoffset为显示设备的黒位电平black level，理论上来说，当屏幕显示黑色时，对应的亮度应该为0，但实际应用中，屏幕显示黑色时对应的亮度值并不是0，而是一个比较小的亮度值，即black level为显示设备能够显示的亮度的最小值；在该步骤中，BLoffset为线性空间中的显示亮度。应当理解，每个显示设备的BLoffset可能不一样，BLoffset可以通过对显示设备进行测量得到。
得到处理后的线性显示光信号RtGtBt之后，根据实际显示设备的色彩空间，对线性显示光信号RtGtBt进行色彩空间转换，将RtGtBt信号转换到显示设备的色彩空间上来，得到与显示设备的显示亮度范围和色彩空间均匹配的处理后的视频信号；示例性的，如果显示设备的色彩空间为sRGB空间，则可以直接进行显示，无需进行色彩空间转换；如果显示设备的色彩空间为YUV，则将RtGtBt信号转换成YUV空间上的视频信号YUV1。
在本申请实施例中，将显示光信号的亮度信号转换到非线性空间，并在非线性空间上对显示亮度进行亮度映射，可以将HDR视频信号的显示亮度范围合理映射到显示设备能够显示的亮度范围内，提升画面的对比度、亮度、细节表现；由于亮度映射是在非线性空间中进行的，由亮度映射引入的误差均匀分布，对视频信号最终的显示效果影响较小，尤其在低亮度显示的情况下映射后的显示亮度分布合理、画面显示不会偏暗。同时，本申请实施例考虑了显示设备的黑位电平对亮度映射曲线的影响，保留了低亮度部分的亮度细节。进一步的，本申请实施例提供的HDR信号亮度处理方法，可以将HDR信号转换成SDR信号，提升SDR显示设备对HDR信号的兼容性。
如图9所示，为本申请实施例提供的一种计算亮度映射曲线TM_Curve的方法流程图。应当理解，图9中所示的方法可用来计算步骤500中的亮度映射曲线TM_Curve，500中的亮度映射曲线TM_Curve还可以通过其他方式计算得到，并不局限于图9所示的方法。
具体的，该计算亮度映射曲线TM_Curve的方法可以包括：
900、获取源及显示设备的亮度范围：
源最大亮度MaxSrcLuma，源最小亮度MinSrcLuma，显示设备最大亮度MaxDispLuma，显示设备最小亮度MinDispLuma，单位均为nits；源最大/最小亮度可根据实际情况及经验确定，也可从HDR信号携带的元数据信息中获取，比如将主控监视器的最大/最小亮度作为源的最大/最小亮度；显示设备最大/最小亮度根据实际显示设备测得，可选的，特殊应用场景下无法测量实际显示设备的亮度值时，可根据经验设定。
901、将源以及显示设备的亮度信号转换到非线性空间。
示例性的，可以选取与503中相同的非线性空间PQ EOTF -1曲线，计算得到非线性空间PQ空间上的非线性源最大亮度maxSL、非线性源最小亮度minSL、非线性显示最大亮度maxDL、非线性显示最小亮度minDL；
maxSL＝PQ_EOTF -1(MaxSrcLuma),minSL＝PQ_EOTF -1(MinSrcLuma),
maxDL＝PQ_EOTF -1(MaxDispLuma),minDL＝PQ_EOTF -1(MinDispLuma)
应当理解，步骤901可以认为是将亮度范围为[0,10000]nits的线性空间的亮度信号映射到[0,1]范围内的电信号。
902、确定KP；
KP为亮度映射曲线的拐点KneePoint，亮度值低于KneePoint的亮度不进行压缩，亮度值高于KneePoint的亮度进行压缩；示例性的，KP在(minDL,maxDL)中选取，根据实际效果可以调节KP的取值；在一种可选的情况中，KP的选取与源亮度和显示设备的显示亮度的差值有关，当源亮度小于等于显示亮度时，无需对源亮度进行压缩，此时，KP＝maxSL；当源亮度远远大于显示亮度，需要压缩的亮度区间较大，KP选取较小的值；当源亮度大于显示亮度且两者相差不大，则KP的值可以取较大值。
903、TM_Curve曲线分段；
根据KP和maxSL把TM_Curve曲线分成三段，e<KP，KP<＝e<＝maxSL，e>maxSL；e为TM_Curve曲线的输入，e＝[0,1]，代表PQ编码的[0,10000]nits；或者说e表示将亮度范围为0-10000nits的亮度信号经过PQ编码转换到非线性空间内范围为[0,1]的电信号。
904、计算e<KP段曲线；
当e<KP时，e1＝e；
905、计算KP<＝e<＝maxSL段曲线；
当KP<＝e<＝maxSL时，e1＝Hermite interpolation(e)＝hmt(e)；
Hermite interpolation采用三次二重Hermite插值，应当理解，插值方法也可采用其他插值方法，公式如下所示：
hmt(x)＝y 0×α 0(x)+y 1×α 1(x)+y′ 0β 0(x)+y′ 1β 1(x)   (32)
Figure PCTCN2019090566-appb-000027
其中x是自变量，其他参数值为：x0＝KP，x1＝maxSL，y0＝KP，y1＝maxDL，y0′＝1； y1′＝0。
y0＝x0表示在第一拐点下的亮度值不压缩，采用线性映射；x1＝maxSL,y1＝maxDL表示把源的最大亮度映射到屏的最大亮度maxDL；x0＝KP,y0'＝1保证曲线在KP这个拐点斜率为1，x1＝maxSL,y1'＝0保证曲线在maxSL这个拐点斜率为0，这样可以保证拟合得到的第二段曲线在两个拐点处是平滑的。
906、计算e>maxSL段曲线；
当e>maxSL时，e1＝maxDL；
907、得到亮度映射曲线e1；
根据904、905、906得到曲线e1；
908、计算BlackLevelLift；
由于显示设备有最小亮度，即使给显示设备一个纯黑信号，在显示设备上也会显示为最小亮度值，这个最小亮度称为黒位电平black level。因此需要对显示亮度进行一定的黒位电平抬升BlackLevelLift，简写为BLL，以保护亮度小于黒位电平的细节亮度；本实施例中黒位电平抬升BLL值根据亮度映射曲线TM_Curve的输入e进行计算：
BLL＝minDL*(1-e)^n   (34)
n>＝1，示例性的，本申请实施例选取n＝5；
应当理解，这里的黑位电平抬升BLL为非线性空间的亮度值，而前述507中的BLoffset是线性空间的亮度值为保证亮度映射曲线TM_Curve能够把源的最小亮度映射到屏的最小亮度，同时黒位电平抬升BLL对TM_Curve的改变较小，需要对BLL进行归一化，得到归一化黒位电平抬升BLLnorm；归一化结果如下所示：
BLLnorm＝MAX((BLL-norm_y1)/(norm_y0-norm_y1)*(minDL-maxSL),0)   (35)
其中，norm_x0＝minSL；norm_y0＝minDL*(1-minSL)^n；
norm_x1＝KP；norm_y1＝minDL*(1-KP)^n；
909、计算输出的亮度映射曲线e2：
亮度映射曲线TM_Curve输出e2＝e1+BLLnorm，其中，BLLnorm为归一化之后的由黑位电平导致的亮度值的提升；
910、转换成一维查找表TM_Curve；
在本实施例中，为了使亮度映射曲线具有普适性，采用一维查找表来表示TM_Curve，因此将映射曲线e2选取n个采样点{(x0,y0),(x1,y1)…(xn,yn)}转换成一维查找表TM_Curve＝{TM_Curve_x,TM_Curve_y}；
TM_Curve_x＝(x0,x1,…xn)是n个采样点的横坐标，即曲线的输入，代表亮度映射前的亮度信号；
TM_Curve_y＝(y0,y1,…yn)是n个采样点的纵坐标，即曲线的输出，代表亮度映射后的亮度信号；
实际使用可根据需求采用其他方式(例如公式化)来表示映射曲线TM_Curve；具体可参考500中对TM_Curve的描述。
在一种可选的方式中，先按照源信号的亮度范围(minSL，maxSL)对输入亮度进行归一化，然后映射到归一化输出亮度，最后利用显示设备的亮度范围(minDL，maxDL)对归一化输出亮度进行反归一化，得到最终的输出亮度，其亮度映射曲线是一条相对值映射曲线，映射曲线输入[0,1]表示[minSL，maxSL]，映射曲线输出[0,1]表示[minDL，maxDL]；而本申请实施例中亮度映射曲线直接根据输入亮度计算输出亮度，其亮度映射曲线是一条绝对值映射曲线，映射曲线输入[0,1]表示[0,10000]nits，映射曲线输出[0,1]表示[0,10000]nits。
本申请实施例通过选取两个阈值，将传统的亮度映射曲线分成三部分，低亮度部分(即低于第一阈值的亮度部分)不进行压缩，高亮度部分(即大于第一阈值的亮度部分)进行 压缩，而且高亮度部分又进一步分成了两个部分，其中大于第一阈值且小于第二阈值的亮度部分通过曲线进行柔性压缩，该曲线是基于第一阈值和第二阈值拟合得到的，尽可能保留该部分亮度的细节；大于第二亮度拐点的部分进行第二压缩，即大于第二阈值的亮度都映射到第二阈值上；对亮度进行分段映射，充分考虑了各个亮度的特性，尽可能保留亮度细节，提升亮度映射的合理性。同时，该亮度映射曲线考虑了显示设备的黑位电平对亮度映射曲线的影响，保留了低亮度部分的亮度细节。
如图10所示，为本申请实施例提供的另一种HDR视频信号亮度处理的方法流程图。
示例性的，本申请实施例中输入的源视频信号为场景光信号HDR HLG信号。
该方法可以包括：
1000、计算亮度映射曲线TM_Curve。
同步骤600，具体请参考步骤600处的描述。该亮度映射曲线可以是图9中所示的方法计算得到的亮度映射曲线，也可以是其他方法获得的亮度映射曲线，本申请实施例对此不作限定。示例性的，当基于图9中所示的方法计算亮度映射曲线时，可以设置MaxSrcLuma＝L W,MinSrcLuma＝0,L W为HLG场景信号转换成显示信号后对应的最大显示亮度。
1001、计算线性场景光信号RsGsBs；
示例性的，本申请实施例中输入的视频信号为YUV空间的HDR HLG信号YUV0，经过色彩空间转换，得到RGB色彩空间中的非线性场景光信号R’sG’sB’s；进一步的，将非线性场景光信号R’sG’sB’s转换成线性场景光信号RsGsBs，在一种可选的情况中，基于HLG OETF-1曲线将非线性场景光信号R’sG’sB’s转换成线性场景光信号RsGsBs。色彩空间转换采用标准的色彩空间转换过程；其中Rs、Gs、Bs＝[0,1]；在一种可选的情况中，也可以处理其他YCC空间的HDR HLG信号；
应当理解，在这种情况中，认为YUV转换至RGB属于色彩空间转换，而非线性RGB转线性RGB信号并没有进行色彩空间的转换，即在这种情况中，非线性RGB信号和线性RGB信号同属RGB色彩空间。
应当理解，在一种可选的情况中，也可以认为将输入的视频信号从YUV信号转换为非线性场景光信号R’sG’sB’s为一次色彩空间转换，将非线性场景光信号R’dG’sB’s转换为线性场景光信号RsGsBs也可以认为是一次色彩空间转换，在这种情况中，非线性RGB场景光信号和线性RGB场景光信号属于不同的色彩空间。
1002、计算场景光亮度Ys；
根据以下公式计算场景光亮度Ys：
Ys＝cr*Rs+cg*Gs+cb*Bs   (36)
Ys为线性的场景亮度信号，场景亮度信号为表征场景光信号的亮度分量；其中，参数cr，cg，cb的选择与线性场景光信号RsGsBs的色域有关，当线性场景光信号RsGsBs的色域为BT.2020时，计算参数cr＝0.2627，cg＝0.6780，cb＝0.0593；当线性场景光信号的色域不同时，对应的计算参数也会不同，因此计算亮度时需要根据光信号所处的色域选取对应色域下线性亮度计算参数；计算得到的场景亮度的范围为Ys＝[0,1]；
在一种可选的情况中，利用ITU BT.2100中定义的HLG OOTF将线性场景光信号HLG转换成线性显示光信号后进行显示，该方法根据实际显示设备的最大亮度和最小亮度，计算出“系统伽马”相关参数将线性场景光信号转换成显示设备动态范围内的线性显示光信号。
HLG OOTF定义如下：
Figure PCTCN2019090566-appb-000028
上述公式将线性场景光信号R SG SB S转换成线性显示光信号R DG DB D；其中Y S为线性场景光亮度，α、β、γ为计算参数，计算如下；
Y S＝0.2627R S+0.6780G S+0.0593B S
α＝(L W-L B)
β＝L B                (38)
γ＝1.2+0.42Log 10(L W/1000)    (39)
其中L W，L B分别为显示设备的最大亮度、最小亮度。
在一种可选的情况中，得到场景亮度Y S之后，转换成显示亮度，并对显示亮度进行亮度映射处理，或者说进行tone mapping处理。
1003、将场景亮度转换成线性显示亮度Yd；
应当理解，基于场景光信号得到的亮度是场景光亮度信号，需要将场景光亮度信号转换为显示光亮度信号，该步骤是亮度信号类型转换的一种示例性的实现方式，利用系统伽马将线性场景亮度Ys转换成线性显示亮度Yd；
Yd＝(Ys^γ)*L W   (40)
其中γ＝1.2+0.42Log10(L W/1000)；
示例性的，本申请实施例中设置显示信号最大亮度LW＝1000nits；LW也可设置其他值。
将场景亮度转换为显示亮度提升了对场景信号和显示信号的处理的兼容性。
1004、将显示亮度Yd转换到非线性空间NL_Yd；
同步骤603，请参考步骤603部分的描述，此处不再赘述。
1005、对非线性显示光亮度信号NL_Yd进行亮度映射；
同步骤604，请参考步骤604部分的描述，此处不再赘述。
1006、将映射后的非线性显示光亮度信号NL_Yt转换到线性空间，得到线性显示光亮度信号Yt；
同步骤605，请参考步骤605部分的描述，此处不再赘述。
1007、计算线性亮度增益K；
示例性的，K＝Yt/Ys，即K表征在线性显示空间中亮度映射后的亮度信号Yt和亮度映射前的场景亮度信号Ys的比值；
1008、计算处理后的线性显示光信号；
(Rt,Gt,Bt)＝K*(Rs,Gs,Bs)+(BLoffset,BLoffset,BLoffset)    (41)
Rt、Gt、Bt分别为亮度映射处理后的线性显示光信号的红色分量、绿色分量和蓝色分量；BLoffset为显示设备的黒位电平black level，具体请参考步骤507部分的描述。
得到处理后的线性显示光信号RtGtBt之后，根据实际显示设备的色彩空间，对线性显示光信号RtGtBt进行色彩空间转换，将RtGtBt信号转换到显示设备的色彩空间上来，得到与显示设备的显示亮度范围和色彩空间均匹配的处理后的视频信号；示例性的，如果显示设备的色彩空间为sRGB空间，则可以直接进行显示，无需进行色彩空间转换；如果显 示设备的色彩空间为YUV，则将RtGtBt信号转换成YUV空间上的视频信号YUV1。
在本申请实施例中，将场景光信号的亮度信号转换成显示光亮度信号，并将线性显示光亮度信号转换到非线性空间，并在非线性空间上对显示亮度进行亮度映射，可以将HDR视频信号的显示亮度范围合理映射到显示设备能够显示的亮度范围内，提升画面的对比度、亮度、细节表现；将HLG场景信号转换成线性显示光亮度信号之后，不直接进行显示，而是再将线性显示光亮度信号转换到非线性空间中，在非线性空间中进行亮度映射，由于亮度映射是在非线性空间中进行的，由亮度映射引入的误差均匀分布，对视频信号最终的显示效果影响较小，尤其在低亮度显示的情况下映射后的显示亮度分布合理、画面显示不会偏暗同时，本申请实施例考虑了显示设备的黑位电平对亮度映射曲线的影响，保留了低亮度部分的亮度细节。进一步的，本申请实施例提供的HDR信号亮度处理方法，可以将HDR信号转换成SDR信号，提升SDR显示设备对HDR信号的兼容性。
应当理解，为了便于理解，图6、图9、图10中的方法实施例以步骤的形式对方法进行描述，虽然在方法流程图6、9、10中示出了方法的顺序，但是在某些情况下，可以以不同于此处的顺序执行所描述的步骤。
如图13所示，为本申请实施例提供的一种视频信号处理的装置，该装置包括：亮度获取单元1301、第一转换单元1302、亮度映射单元1303、第二转换单元1304、增益计算单元1305以及显示信号获取单元1306，可选的，该装置还可以包括补偿单元1307和色彩空间转换单元1308。
亮度获取单元1301，用于获得第一线性亮度信号，该第一线性亮度信号基于待处理视频信号对应的第一线性红绿蓝RGB信号获得。具体请参考步骤501部分的描述，此处不再赘述。
可选的，当待处理的视频信号为HDR PQ信号时，亮度获取单元1301，具体用于对该PQ信号进行色彩空间转换，以获得第一非线性RGB信号；根据PQ电光转移函数，将该第一非线性RGB信号转换为该第一线性RGB信号；基于该第一线性RGB信号的各基色信号进行计算，以获得该第一线性亮度信号。具体请参考步骤601、602部分的描述。
当待处理视频信号为混合对数伽玛HLG信号，该亮度获取单元，具体用于：对该HLG信号进行色彩空间转换，以获得第二非线性RGB信号；根据HLG光电转移逆函数，将该第二非线性RGB信号转换为第二线性RGB信号；基于该第二线性RGB信号的各基色信号进行计算，以获得第三线性亮度信号；对该第三线性亮度信号进行亮度信号类型转换，以获得该第一线性亮度信号。具体请参考步骤1001、1002、1003部分的描述。
第一转换单元1302，用于将该第一线性亮度信号转换为第一非线性亮度信号；
具体请参考步骤502部分的描述，此处不再赘述。可选的，第一转换单元1302还可以用于完成步骤603、1004；
亮度映射单元1303，用于对该第一非线性亮度信号进行分段亮度映射，以得到第二非线性亮度信号；
具体请参考步骤503部分的描述，此处不再赘述。可选的，亮度映射单元1303还可以用于完成步骤604、1005；应当理解，亮度映射单元可以调用存储在存储器中的查找表、亮度映射曲线或亮度映射公式对待处理的视频信号完成亮度映射。关于亮度映射曲线请参考600以及图9所对应的实施例部分的描述。
第二转换单元1304，用于将该第二非线性亮度信号转换为第二线性亮度信号；
具体请参考步骤504部分的描述，此处不再赘述。可选的，第二转换单元1304还可以用于完成步骤605、1006；
增益计算单元1305，用于计算该第二线性亮度信号与该第一线性亮度信号的亮度增益；
具体请参考步骤505部分的描述，此处不再赘述。可选的，增益计算单元1305还可以用于完成步骤606、1007；
显示信号获取单元1306，用于基于该亮度增益和该第一线性RGB信号的乘积，获得该待处理视频信号对应的RGB显示信号；
具体请参考步骤506部分的描述，此处不再赘述。
补偿单元1307，用于对该RGB显示信号的各基色值叠加黑位电平提升BlackLevelLift，以获得处理后的RGB显示信号，该BlackLevelLift为该显示设备的显示亮度的最小值；
可选的，显示信号获取单元1306和补偿单元1307可以共同用于完成步骤607、1008；
色彩空间转换单元1308，用于对处理后的RGB显示信号进行色彩空间转换，以获得目标显示信号，其中，该目标显示信号的色彩格式和显示设备对应的色彩格式相同。应当理解，如果显示设备对应的色彩格式为RGB，则可以直接进行显示、不需要进行色彩空间的转换。
示例性的，下面为本申请提供的一种视频信号处理的具体实施方式。该申请实施例将HDR HLG信号转换成SDR信号，以适配SDR TV。在本申请实施例中，先进行饱和度映射，再进行亮度映射和色域映射，在其他可能的实时方式中，可以调换饱和度映射、亮度映射和色域映射的处理顺序，本申请实施例对此不做限定。
设Y`sCbsCrs是终端经过AVS2解码重建和色度上采样恢复成的4:4:4YCbCr非线性视频信号。各分量均为10比特的数字编码值。
1)计算YiCbiCri信号(饱和度映射)；
a.计算归一化原始亮度：
Y norm＝(Y-64)/(940-64)       (42)
Ynorm应Clip到[0,1]范围内；
b.计算饱和度映射增益SmGain
SmGain＝f sm(Y norm)         (43)
其中f sm()为饱和度映射曲线，根据亮度映射曲线f tm()进行计算得到，其计算步骤为：
i.将亮度映射曲线ftm()转换到线性空间上，得到线性亮度映射曲线：
f tmL(L)＝PQ_EOTF(f tm(PQ_EOTF -1(L)))     (44)
其中，L为输入线性亮度，单位nit，f tm(L)的结果为线性亮度，单位nit；
ii.将线性亮度映射曲线f tmL()转换到HLG空间上，得到HLG信号上的亮度映射曲线：
Figure PCTCN2019090566-appb-000029
其中e为归一化HLG信号亮度，f tmHLG(e)结果为归一化HLG信号亮度；
iii.计算饱和度映射曲线f sm()
Figure PCTCN2019090566-appb-000030
其中饱和度映射曲线输入e，f sm(e)为HLG空间上的饱和度映射增益；
c.计算饱和度映射后信号
Figure PCTCN2019090566-appb-000031
YiCbiCri信号是10比特限制范围数字编码值，其中Yi数值应该在[64,940]区间内，而Cbi，Cri数值应该在[64,960]区间内。
2)计算非线性R`sG`sB`s信号；
Figure PCTCN2019090566-appb-000032
Figure PCTCN2019090566-appb-000033
其中的Y` sCb sCr s信号是10比特限制范围数字编码值，经过该处理得到的R` sG` sB` s是浮点非线性基色值，数值应clip到[0,1]区间的。
2)计算线性R sG sB s信号，并计算输入信号线性亮度Y s；
E s＝HLG_OETF -1(E` s)       (50)
等式中的E s表示R sG sB s信号中任一分量的线性基色值，其数值在[0,1]区间；E` s指R` sG` sB` s信号中任一分量的非线性基色值。函数HLG_OETF -1()根据ITU BT.2100定义如下：
Figure PCTCN2019090566-appb-000034
其中a＝0.17883277,b＝1-4a,c＝0.5-a*ln(4a)
线性亮度Y s计算如下：
Y s＝0.2627R s+0.6780G s+0.0593B s       (52)
Y s是实数，其数值在[0,1]区间。
3)计算Y t信号；
a.计算显示亮度Y d：
Y d＝1000(Y s) 1.2       (53)
b.计算视觉线性亮度Y dPQ：
Y dPQ＝PQ_EOTF -1(Y d)       (54)
其中，
Figure PCTCN2019090566-appb-000035
m 1＝2610/16384＝0.1593017578125
m 2＝2523/4096*128＝78.84375
c 1＝3424/4096＝0.8359375＝c 3-c 2+1
c 2＝2413/4096*32＝18.8515625
c 3＝2392/4096*32＝18.6875
c.进行亮度映射得到Y tPQ：
Y tPQ＝f tm(Y dPQ)        (56)
等式中的f tm()定义如下：
Figure PCTCN2019090566-appb-000036
其中函数hmt()定义如下：
hmt(x)＝0.4064×α 0(x)+0.5791×α 1(x)+β 0(x)
Figure PCTCN2019090566-appb-000037
d.计算归一化亮度映射后线性亮度Y t：
Y t＝PQ_EOTF(Y tPQ)          (59)
其中，
Figure PCTCN2019090566-appb-000038
因此，Yt的计算公式为：
Y t＝PQ_EOTF(f tm(PQ_EOTF -1(1000(Y s) 1.2))       (61)
Y t是实数，其数值应clip到[0,200]区间。
4)计算亮度映射增益TmGain；
亮度映射增益TmGain的计算如下面等式所示：
Figure PCTCN2019090566-appb-000039
5)计算R tmG tmB tm信号；
E tm＝E s×TmGain        (63)
等式中Es表示R sG sB s信号中任一分量，E tm表示R tmG tmB tm信号中任一分量。
6)计算R tG tB t信号(色域映射)；
Figure PCTCN2019090566-appb-000040
经过该处理得到的RtGtBt是浮点线性基色值，数值应clip到[0,200]区间的。
7)计算R` tG` tB` t信号；
E` t＝(E t/200) 1/γ      (65)
8)计算Y tCb tCr t信号；
Figure PCTCN2019090566-appb-000041
Figure PCTCN2019090566-appb-000042
示例性的，该实施例中的γ可以取2.2或2.4，也可以取其他数值，γ的取值可以根据实际情况选取，本申请实施例对此不做限定。
其中的R`tG`tB`t是的非线性基色值，数值在[0,1]区间。经过该处理得到的Y`tCbtCrt信号是10比特限制范围数字编码值，其中Y`t数值应该在[64,940]区间内，而Cbt，Crt数值应该在[64,960]区间内。
如图14所示，为本申请提供的一种HDR终端技术方案处理流程图，示出了在整个视频端到端系统中的技术框架和涉及范围。由图14所示，本申请实施例提供的端到端系统的技术框架可以将HDR HLG视频信号转换为SDR信号，从而适配SDR TV，也可以将HDR HLG信号转换为HDR PQ信号，适配HDR PQ TV进行显示；可选的，也可以将HDR PQ信号转换为SDR信号，适配SDR TV进行播放；可选的，还可以丢弃视频信号的动态元数据，仅保留静态元数据，从而适配HDR PQ TV进行播放。应当理解，该技术框架将HDR视频信号及其动态元数据或静态元数据进行第二代音视频编码标准(2 nd Audio Video coding Standard,AVS2)编码成AVS2码流传输给终端，可选的，也可以采用高效率视频编码(High Efficiency Video Coding，HEVC)等其他现有的或者未来可以支持HDR视频的其他新编码标准等，本申请实施例对此并不限定。
下面为本申请实施例给出的一种视频信号转换处理的实施方式。
1101、HDR输入源信号格式
本申请实施例支持的HDR输入信号源，是在终端接收到的AVS2码流。经过解码后，获得10bit 4:2:0格式的整数型YCbCr色差信号(后文简称为YCC信号，本文不再敷述)以及相关的HDR静态或动态元数据。
上述YCC HDR信号可能是表1中列出的几种格式：
表1
Figure PCTCN2019090566-appb-000043
作为参考，在AVS2码流中对4K超高清电视节目视频参数的封装如表2所示：
表2
Figure PCTCN2019090566-appb-000044
Figure PCTCN2019090566-appb-000045
1102、HDR信号输出对接的TV类型以及建议适配方式
终端可能对接的TV由于生产时间的差异，对HDR信号的支持能力有很大差别。通常旧的TV型号难以支持新的信号格式。为此，解码终端需要根据TV对HDR信号支持能力的差异，进行兼容适配。
根据TV对于HDR信号的接收和处理能力不同，终端设备可能需要对接的TV分为以下几类，如表3所示：
表3
电视机类型 色域 转换曲线
类型1 BT.709 伽马
类型2 BT.2020 伽马
类型3 BT.2020 支持PQ
类型4 BT.2020 支持PQ和HLG
本申请实施例根据终端接收到的不同HDR信号格式与对接TV不同的HDR信号支持能力，进行相应的信号转换处理，从而实现信号与TV之间的兼容适配。
1103、HDR HLG信号到SDR TV的兼容适配
设YsCbsCrs是终端经过AVS2解码重建和色度上采样恢复成的4:4:4YCbCr非线性视频信号。各分量均为10bit的数字编码值。
1)计算非线性R`sG`sB`s信号；
Figure PCTCN2019090566-appb-000046
Figure PCTCN2019090566-appb-000047
其中的YsCbsCrs信号是10比特限制范围数字编码值，经过该处理得到的R`sG`sB`s是浮点化的非线性基色值，数值应clip到[0,1]区间的。
输入的HDR HLG视频信号为YCC空间的视频信号，也可以说，HLG为场景光信号，通过色彩空间转换，将HLG视频信号转换到RGB色彩空间，得到非线性的场景光信号R`sG`sB`s。在本申请实施例中，亮度映射的过程在RGB色彩空间中完成。应当理解，在经过该色彩空间转换得到的视频信号为非线性的视频信号，若输入是HLG信号，则经过色彩空间转换得到的是非线性场景光信号。
2)计算线性RsGsBs信号，并计算输入信号亮度Ys；
E s＝HLG_OETF -1(E` s)    (70)
等式中的E s表示R sG sB s信号中任一分量；E` s指R` sG` sB` s信号中任一分量。函数HLG_OETF -1()根据ITU BT.2100定义如下：
Figure PCTCN2019090566-appb-000048
其中，a＝0.17883277,b＝1-4a,c＝0.5-a*ln(4a)5b
基于线性视频信号获得第一线性显示光亮度信号，示例性的，该第一线性显示光亮度信号为线性显示光亮度信号。在本申请实施例中，由于输入的HDR HLG信号，该线性视频信号为线性场景光信号，基于线性场景光信号获得的亮度信号Ys为场景亮度信号。
线性亮度Ys计算如下：
Y s＝0.2627R s+0.6780G s+0.0593B s   (72)
3)计算Yt信号；
a)计算显示亮度Yd：
Y d＝1000(Y s) 1.2    (73)
应当理解，由于输入的信号为HDR HLG信号，为场景光信号，上述步骤中计算得到的Ys为线性场景亮度，亮度映射曲线的输入是显示亮度，因此，在进行亮度映射之前需要先转换成显示光亮度信号Yd。
b)计算非线性亮度Y dPQ：
Y dPQ＝PQ_EOTF -1(Y d)    (74)
其中，
Figure PCTCN2019090566-appb-000049
m 1＝2610/16384＝0.1593017578125
m 2＝2523/4096*128＝78.84375
c 1＝3424/4096＝0.8359375＝c 3-c 2+1
c 2＝2413/4096*32＝18.8515625
c 3＝2392/4096*32＝18.6875
进一步的，亮度映射是在非线性空间中进行的，因此亮度映射输入的是非线性的显示亮度，因此需要将线性显示亮度转换到非线性空间中，得到非线性显示亮度Y dPQ；
c)进行亮度映射得到Y tPQ：
Y tPQ＝f tm(Y dPQ)      (76)
等式中的f tm()定义如下：
Figure PCTCN2019090566-appb-000050
其中函数hmt()定义如下：
hmt(x)＝0.2643×α 0(x)+0.5081×α 1(x)+β 0(x)   (78)
Figure PCTCN2019090566-appb-000051
d)计算归一化亮度映射后线性亮度Yt：
Y t＝PQ_EOTF(Y tPQ)      (80)
其中，
Figure PCTCN2019090566-appb-000052
亮度映射之后得到的信号为非线性的显示光亮度信号，示例性的，可以通过PQ EOTF曲线将非线性的显示光亮度信号转换为线性显示光亮度信号Yt。
综上，Yt的计算公式为：
Y t＝PQ_EOTF(f tm(PQ_EOTF -1(1000(Y s) 1.2))    (82)
其中，Yt是实数，其数值在[0,100]区间。
4)计算亮度映射增益TMgain；
亮度映射增益TmGain的计算如下面等式所示：
Figure PCTCN2019090566-appb-000053
5)计算饱和度映射增益SMgain，示例性的，该实施例中的γ可以取2.2或2.4，也可以取其他数值，γ的取值可以根据实际情况选取，本申请实施例对此不做限定。
a)计算亮度映射前的非线性显示亮度值：
Y dGMM＝(Y d/1000) 1/γ＝(1000(Y s) 1.2/1000) 1/γ    (84)
b)计算亮度映射后的非线性显示亮度值：
Y tGMM＝(Y t/1000) 1/γ   (85)
c)计算饱和度映射增益SmGain
Figure PCTCN2019090566-appb-000054
6)计算RtmGtmBtm信号；
E tm＝E s×TmGain    (87)
等式中E s表示R sG sB s信号中任一分量，E tm表示R tmG tmB tm信号中任一分量。
7)计算RtGtBt信号(色域映射)；
Figure PCTCN2019090566-appb-000055
示例性的，色域映射可以为：
Figure PCTCN2019090566-appb-000056
8)计算R`tG`tB`t信号；
E` t＝EOTF -1(E t)    (90)
根据应用场景不同，函数EOTF -1()可能为BT.1886EOTF的逆函数或BT.2100PQ EOTF曲线的逆函数。示例性的，E` t＝(E t/100) 1/2.4。
9)计算YtCbtCrt信号；
Figure PCTCN2019090566-appb-000057
Figure PCTCN2019090566-appb-000058
其中的R`tG`tB`t是浮点化的非线性基色值，数值在[0,1]区间。经过该处理得到的YtCbtCrt信号是10比特限制范围数字编码值。
10)计算YoCboCro信号(饱和度映射)；
Figure PCTCN2019090566-appb-000059
其中，YoCboCro信号是10比特限制范围数字编码值。本流程适合HDR HLG信号和HDR PQ信号的兼容适配处理。根据不同适配类型，上述处理流程各步参数有所不同。应当理解，在本申请实施例中，对视频信号先进行亮度处理、再进行色域处理，然后再进行饱和度处理。在一种可选的情况中，可以先进行色域处理、亮度处理，再进行饱和度处理。
为了实现HDR HLG信号到上述各类TV的传输，本实施例建议按表4所列方式进行兼容适配。
表4
Figure PCTCN2019090566-appb-000060
表4中，对于类型2电视，通常也具备处理BT.709色域信号的能力，因此本实施例对于类型2电视，也是将HDR HLG信号转为BT.709色域送给类型2电视。
BT.2020色域到BT.709色域转换
本部分处理是HLG信号到SDR信号兼容适配处理的一个环节，由于该处理方法已在BT.2407报告中给予了概念性介绍，因此本节内容引用该ITU报告内容进行资料性说明。
根据ITU报告BT.2407-0 2部分，BT.2020广色域信号到BT.709信号的转换可以采用基于线性矩阵转换的方法来实现。这种方法除了将输出信号做hard-clip外，完全就是ITU标准BT.2087的逆过程。
如图15所示，为转换过程的图示。
具体而言有下列步骤：
(1)非线性到线性信号的转换(N to L)
假设一个归一化的BT2020非线性RGB信号为((E` RE` GE` B)，各个分量信号经过一个转换函数实现到线性信号(E RE GE B)的转换。在本申请实施例中，该转换函数可以为HLG EOTF函数。
(2)矩阵(M)
BT2020线性RGB信号转换为BT709线性RGB信号，可以通过下面的矩阵计算完成：
Figure PCTCN2019090566-appb-000061
(3)线性信号到非线性信号的转换(L to N)
根据ITU-BT.2087-0标准，BT709线性RGB信号(E RE GE B)要用于BT709显示设备，应该使用ITU BT.1886定义的OETF转换为BT709非线性RGB信号(E` RE` GE` B)。示例性的，该实施例中的γ可以取2.2或2.4，也可以取其他数值，γ的取值可以根据实际情况选取，本申请实施例对此不做限定。
公式如下所示：
E＝(E`) γ,0≤E`≤1    (95)
1104、HDR HLG信号到HDR PQ TV的兼容适配处理
根据ITU报告BT.2390-4 7.2部分，首先约定从HLG到PQ信号的参考峰值亮度Lw为1000nit，黑位Lb为0。
根据该报告，采用图16所示的过程，在1000nit以内color volume内，可以生成与HLG图像相同的PQ图像：
HDR HLG信号到HDR PQ信号的转换处理的具体过程为：
(1)1000nit的HLG源信号经过HLG的OETF反函数可生成线性场景光信号；
(2)线性场景光信号经过HLG的OOTF函数可生成线性显示光信号；
(3)线性显示光信号经过PQ的EOTF反函数可生成1000nit的PQ显示光信号；
该处理流程涉及的是HDR HLG信号转换成HDR PQ信号在TV上显示的过程。
该场景下完整的处理流程如下所示：
设YsCbsCrs是终端经过AVS2解码重建和色度上采样恢复成的4:4:4YCbCr非线性视频信号。各分量均为10比特的数字编码值。
1)计算非线性R`sG`sB`s信号；
Figure PCTCN2019090566-appb-000062
Figure PCTCN2019090566-appb-000063
其中的YsCbsCrs信号是10比特限制范围数字编码值，经过该处理得到的R`sG`sB`s是浮点化的非线性基色值，数值应clip到[0,1]区间的。
输入的HDR HLG视频信号为YCC空间的视频信号，也可以说，HLG为场景光信号，通过色彩空间转换，将HLG视频信号转换到RGB色彩空间，得到非线性的场景光信号R`sG`sB`s。在本申请实施例中，亮度映射的过程在RGB色彩空间中完成。应当理解，在经过该色彩空间转换得到的视频信号为非线性的视频信号，若输入是HLG信号，则经过色彩空间转换得到的是非线性场景光信号。
2)计算线性RsGsBs信号，并计算输入信号线性亮度Ys；
将非线性场景光信号R`sG`sB`s转换成线性场景光信号RsGsBs，在一种可选的情况中，可基于HLG光电转移逆函数将非线性场景光信号R`sG`sB`s转换成线性场景光信号RsGsBs。
E s＝HLG_OETF -1(E` s)    (98)
等式中的Es表示RsGsBs信号中任一分量；E`s指R`sG`sB`s信号中任一分量。函数HLG_OETF -1()根据ITU BT.2100定义如下：
Figure PCTCN2019090566-appb-000064
其中a＝0.17883277,b＝1-4a,c＝0.5-a*ln(4a)5b
基于线性视频信号获得第一线性显示光亮度信号，示例性的，该第一线性显示光亮度信号为线性显示光亮度信号。在本申请实施例中，由于输入的HDR HLG信号，该线性视频信号为线性场景光信号，基于线性场景光信号获得的亮度信号Ys为场景亮度信号，在进行亮度映射之前需要先转换成显示光亮度信号Yd。
线性亮度Ys计算如下：
Y s＝0.2627R s+0.6780G s+0.0593B s   (100)
3)计算Yd信号；
Y d＝1000(Y s) 1.2    (101)
4)计算亮度映射增益TmGain
亮度映射增益TmGain的计算如下面等式所示：
Figure PCTCN2019090566-appb-000065
5)计算R tmG tmB tm信号；
E tm＝E s×TmGain    (103)
等式中Es表示RsGsBs信号中任一分量，E tm表示R tmG tmB tm信号中任一分量。
6)计算R`tG`tB`t信号；
E` t＝PQ_EOTF -1(E t)   (104)
式中函数PQ_EOTF -1()，参考ITU BT.2100表4定义如下：
Figure PCTCN2019090566-appb-000066
其中
m1＝2610/16384＝0.1593017578125
m2＝2523/4096*128＝78.84375
c1＝3424/4096＝0.8359375＝c3-c2+1
c2＝2413/4096*32＝18.8515625
c3＝2392/4096*32＝18.6875
7)计算YtCbtCrt信号；
Figure PCTCN2019090566-appb-000067
Figure PCTCN2019090566-appb-000068
其中的R`tG`tB`t是浮点化的非线性基色值，数值在[0,1]区间。经过该处理得到的YtCbtCrt信号是10比特限制范围数字编码值。
示例性的，终端可以根据与对接HDR PQ TV确定的帧率、位宽、色度下采样方式，完成帧率调整、位宽调整及4:4:4到4:2:2/4:2:0下采样等后续处理，可将转换生成的HDR PQ信号传输给HDR PQ TV。
1105、HDR PQ信号的兼容适配
为了实现HDR PQ信号到上述各类TV的传输，本实施例建议按表5所列方式进行兼容适配。
表5
Figure PCTCN2019090566-appb-000069
HDR PQ信号到SDR信号的转换处理
基于1103部分所述的通用过程，可以实现HDR PQ信号到SDR信号的转换处理。示例性的，终端可以根据与对接SDR TV确定的帧率、位宽、色度下采样方式，完成帧率调整、位宽调整及4:4:4到4:2:2/4:2:0下采样等后续处理，即可将转换生成的SDR信号传输给SDR TV。
HDR PQ信号送到HDR PQ TV的适配
考虑到目前支持HDR PQ信号的HDR TV均具备HDMI2.0A或以上的HDMI接口能力，且有较为成熟的HDR PQ信号处理能力，本实施例建议当接收到HDR PQ信号，应该将HDR PQ信号以及静态元数据通过HDMI2.0A及以上版本的HDMI接口，直接输出给HDR PQ TV，由TV自行完成HDR PQ信号的后续显示处理。至于动态元数据，则由于HDMI接口限制无法传递给TV，应在解码之后做丢弃处理。
下面为本申请实施例提供的一种测试方案。
具体做法是将待评价图像和参考图像按后述各测试方案持续播放一定时间给观察者，然后在播放后留出一定的时间间隔供观察者打分，最后将所有给出的分数取平均作为该序列的评价值，即该待评图像的评价值。
参考BT.500-13对图像质量和损伤的打分规则，由多位测试者对各种测试对象按照表6的打分标准进行打分。
表6
Figure PCTCN2019090566-appb-000070
Figure PCTCN2019090566-appb-000071
HDR HLG信号适配SDR TV
测试目的：通过对比测试说明HDR HLG信号送往SDR TV时，HLG到SDR的图像适配转换是否能够提供有益的图像效果。如图17所示为一种测试组网的示意图。其中：
DUT1:HLG转SDR 709
DUT2:HLG转SDR BT2020
BenchMark1:HLG在SDR BT709观看
BenchMark1:HLG在HDR HLG TV以HLG BT2020模式观看
HDR HLG信号适配HDR PQ TV
如图18所示，为本申请实施例提供的另一种测试方案的组网方式示意图。
DUT:HLG转HDR PQ BT2020
Benchmark1:HLG在HDR PQ TV以SDR BT709模式观看
Benchmark2:HLG在HDR PQ TV以HLG BT2020模式观看
对PQ曲线HDR视频，HDR端到端系统见图19所示。通过节目制作，获得PQ曲线HDR视频和静态元数据，HDR视频制作参数符合GY/T 315-2018标准的规定。HDR前处理实现动态元数据的提取，得到用于编码传输的HDR视频和元数据，经过AVS2编码与封装后，在网络中进行传输。在接收端，AVS2解码器解码后得到HDR视频和元数据。对于SDR显示终端，利用HDR视频和元数据重构得到SDR视频进行显示；对于HDR显示终端，若终端显示能力与制作传输的HDR视频亮度相同，则直接进行HDR显示；若终端显示能力与制作传输的HDR视频亮度不同，则利用HDR视频和元数据根据终端显示能力适配后显示。
对HLG曲线HDR视频，当节目制作采用最高亮度为1000cd/m2时，HDR端到端系统见图20所示。通过节目制作，获得HLG曲线HDR视频，HDR视频制作参数符合GY/T 315-2018标准的规定。对HDR视频采用AVS2编码后，在网络中进行传输。在接收端，AVS2解码器解码后得到HDR视频，直接在SDR与HDR终端显示。
对HLG曲线HDR视频，当节目制作采用最高亮度不为1000cd/m2时，HDR端到端系统见图21所示。通过节目制作，获得HLG曲线HDR视频和静态元数据，HDR视频制作参数符合GY/T 315-2018标准的规定。对HDR视频和静态元数据采用AVS2编码和封装后，在网络中进行传输。在接收端，AVS2解码器解码后得到HDR视频，AVS2解码器解码后得到HDR视频和静态元数据。对于SDR显示终端，直接显示；对于HDR显示终端，可利用静态元数据采用GY/T 315-2018表5中规定的方法对显示伽马值进行调整后显示。
元数据信息
元数据信息metadata_info()包含了动态元数据，当它与关联的所传图像相结合时可以重构HDR和图像。
HDR图像特性
hdr_characteristics()包含了HDR图像信号特征，即HDR图像颜色空间的标识,以及用于调校HDR图像监视器基色的标识。
HDR图像颜色空间hdrPicColourSpace
此变量表示符合CIE1931中定义的HDR图像颜色空间的基色和基准白坐标。
HDR主监视器颜色空间hdrDisplayColourSpace
此变量表示符合CIE1931定义的调校HDR图像的主监视器颜色空间的基色和基准白坐标。
HDR主监视器最大显示亮度hdrDisplayMaxLuminance
此变量定义了对HDR图像进行调校的主监视器的标称最大显示亮度，以坎德拉每平方米(cd/m2)为单位且四舍五入到50cd/m2的整数倍。
HDR主监视器最小显示亮度hdrDisplayMinLuminance
此变量定义了对HDR图像进行调校的主监视器的标称最小显示亮度。以0.0001cd/m2为单位。
hdrDisplayMinLuminance应小于hdrDisplayMaxLuminance。如果变量值未知，建议设置为0。
SDR图像特性
sdr_characteristics()包含了SDR图像信号特征，即SDR图像的颜色空间标识，以及用于调校SDR图像主监视器的标称最大和最小亮度值。
SDR图像颜色空间sdrPicColourSpace
表示符合CIE1931定义的SDR图像颜色空间的基色和基准白坐标。
SDR主监视器最大亮度sdrDisplayMaxLuminance
该变量定义了用于调校SDR图像的主监视器的标称最大显示亮度，以1cd/m2为单位，且四舍五入到50cd/m2的整数倍。
SDR主监视器最小亮度sdrDisplayMinLuminance
luminance_mapping_variables()包含了亮度映射变量，用于构建查找表lutMapY。
色调映射输入信号黑电平偏移tmInputSignalBlackLevelOffset
在亮度映射曲线重构过程中，此变量用于计算信号增益的第一步,表示在色度体积重构过程中消去的黑电平偏移。该变量值应在[0,1]范围内，且为(1÷255)的整数倍。
接收终端
规定不同应用场景下，接收终端如何实现本申请实施例规定的HDR后处理，分为以下两种情况：
(1)机顶盒加电视机，后处理模块在机顶盒
用户根据电视机的显示能力，如最大显示亮度、最小显示亮度、色域等，并利用机顶盒的遥控器进行参数配置，机顶盒获知电视机显示能力。机顶盒对接收到的节目码流进行分析，得到节目是SDR节目/HDR节目、最大亮度、最小亮度等信息，若与电视机的显示能力相匹配，对音视频解码并通过HDMI送到电视进行显示；若与电视机的现实能力不匹配，对音视频解码并通过显示适配后获得与电视机显示能力适配的节目信号，通过HDMI送到电视进行显示。
(2)一体机，后处理模块在电视
一体机对接收到的节目码流解码，根据电视机的实际显示能力进行调整适配和显示。
本申请实施例描述了在HDMI接中传输HDR元数据的方式，其中，HDMI2.0a规范规定了如何在HDMI接口中传输HDR静态元数据，HDMI2.1规范规定了如何在 HDMI接口中传输HDR动态元数据。
HDMI接口中传输HDR静态元数据
HDMI 2.0a标准采用了CEA-861.3-2014对HDR静态元数据传输的规范，当上游源处理设备(例如机顶盒)无法处理HDR静态元数据时，HDR静态元数据可在CEA-861.3-2014接口中传输给能接收处理HDR静态元数据的接收端(例如电视机)。本附录补充说明了CEA-861.3-2014中HDR静态元数据数据块“HDR Static Metadata Data Block”(C.2.2，对应CEA-861.3-2014规范的4.2)和动态范围主监信息帧“Dynamic Metadata and Mastering InfoFrame”(C.2.3，对应CEA-861.3-2014规范的3.2)的信息传输和映射。
HDR静态元数据数据块
HDR静态元数据数据块“HDR Static Metadata Data Block”用于将接收端HDR静态元数据支持能力信息传输给源设备。当接收端传输给源设备的该数据块中的“ET_2”位等于1，表示接收端支持GY/T 315-2018规定的PQ EOTF曲线，“SM_0”位等于1，表示支持本文规定的静态元数据。
动态范围主监信息帧
动态范围主监信息帧“Dynamic Range and Mastering InfoFrame”由源设备用于标识并传输HDR动态元数据给接收设备。
源设备利用“EOTF”值为2来标记传输流的EOTF为GY/T 315-2018规定的PQ曲线，“Static_MetaData_Descriptor_ID”为0来标记本标准的静态元数据在用户扩展信息中携带,从而传输符合本标准规范的静态元数据。
HDMI接口中传输HDR动态元数据
HDMI 2.1标准采用了CTA-861-G-2016对HDR动态元数据传输的规范，CTA-861-G2016规定了如何在CE终端接口中承载HDR元数据。当上游源处理设备(例如机顶盒)无法处理HDR元数据时，HDR元数据可在CTA-861-G-2016接口中传输给能接收处理HDR元数据的接收端(例如电视机)。本附录补充说明了CTA-861-G-2016中HDR动态元数据数据块“HDR Dynamic Metadata Data Block”(C.3.2，对应CTA-861-G-2016规范的7.5.14)和HDR动态元数据扩展信息帧“HDR Dynamic Metadata Extended InfoFrame”(C.3.3，对应CTA-861-G-2016规范的6.10.1)的信息传输和映射。
HDR动态元数据数据块
HDR动态元数据数据块“HDR Dynamic Metadata Data Block”用于将接收端HDR动态元数据支持能力信息传输给源设备。当接收端传输给源设备的该数据块中的支持动态元数据类型“Supported HDR Dynamic Metadata Type”的值为0x0002时，表示该接收端支持本标准规范的动态元数据。
在支持动态元数据类型“Supported HDR Dynamic Metadata Type”的值为0x0002时，数据块中支持标记“Support Flags”字节代表了不同的HDR传输模式。其0～3位组成的二进制值大于等于1，第5位等于1,表示支持本标准规范的HDR传输模式。
HDR动态元数据扩展信息帧
HDR动态元数据扩展信息帧“HDR Dynamic Metadata Extended InfoFrame”由源设备用于标识并传输HDR动态元数据给接收设备。
源设备利用扩展信息帧类型“Extended InfoFrame Type”值为0x0002来标记本标准的 动态元数据在用户扩展信息中携带,从而传输符合本标准规范的动态元数据。
本申请实施例还提供了一种计算机可读存储介质，该计算机可读存储介质中存储有指令，当其在计算机上运行时，使得计算机执行上述任一个方法中的一个或多个步骤。上述信号处理装置的各组成模块如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在所述计算机可读取存储介质中。
基于这样的理解，本申请实施例还提供一种包含指令的计算机程序产品，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)或其中的处理器执行本申请各个实施例所述方法的全部或部分步骤。该存储介质的种类请参考存储器302的相关描述。
以上所述实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的范围。例如，装置实施例中的一些具体操作可以参考之前的方法实施例。

Claims (25)

  1. 一种视频信号的处理方法，其特征在于，所述方法包括：
    获得第一线性亮度信号，所述第一线性亮度信号基于待处理视频信号对应的第一线性红绿蓝RGB信号获得；
    将所述第一线性亮度信号转换为第一非线性亮度信号；
    对所述第一非线性亮度信号进行分段亮度映射，以得到第二非线性亮度信号；
    将所述第二非线性亮度信号转换为第二线性亮度信号；
    计算所述第二线性亮度信号与所述第一线性亮度信号的亮度增益；
    基于所述亮度增益和所述第一线性RGB信号的乘积，获得所述待处理视频信号对应的RGB显示信号。
  2. 根据权利要求1所述的方法，其特征在于，所述待处理视频信号为感知量化PQ信号，所述获得第一线性亮度信号，包括：
    对所述PQ信号进行色彩空间转换，以获得第一非线性RGB信号；
    根据PQ电光转移函数，将所述第一非线性RGB信号转换为所述第一线性RGB信号；
    基于所述第一线性RGB信号的各基色信号进行计算，以获得所述第一线性亮度信号。
  3. 根据权利要求1所述的方法，其特征在于，所述待处理视频信号为混合对数伽玛HLG信号，所述获得第一线性亮度信号，包括：
    对所述HLG信号进行色彩空间转换，以获得第二非线性RGB信号；
    根据HLG光电转移逆函数，将所述第二非线性RGB信号转换为第二线性RGB信号；
    基于所述第二线性RGB信号的各基色信号进行计算，以获得第三线性亮度信号；
    对所述第三线性亮度信号进行亮度信号类型转换，以获得所述第一线性亮度信号。
  4. 根据权利要求1至3任一项所述的方法，其特征在于，在所述获得所述待处理视频信号对应的RGB显示信号之后，还包括：
    对所述RGB显示信号进行色彩空间转换，以获得目标显示信号，其中，所述目标显示信号的色彩格式和显示设备对应的色彩格式相同。
  5. 根据权利要求4所述的方法，其特征在于，在所述获得所述待处理视频信号对应的RGB显示信号之后，还包括：
    对所述RGB显示信号的各基色值叠加黑位电平提升BlackLevelLift，以获得处理后的RGB显示信号，所述BlackLevelLift为所述显示设备的显示亮度的最小值；
    对应的，所述对所述RGB显示信号进行色彩空间转换，包括：
    对所述处理后的RGB显示信号进行色彩空间转换。
  6. 根据权利要求4或5任一项所述的方法，其特征在于，所述对所述第一非线性亮度信号进行分段亮度映射，以得到第二非线性亮度信号，包括：
    确定第一阈值和第二阈值，所述第一阈值小于所述第二阈值；
    当所述第一非线性亮度信号的亮度值小于或等于所述第一阈值时，所述第二非线性亮度信号的亮度值等于所述第一非线性亮度信号的亮度值；
    当所述第一非线性亮度信号的亮度值大于所述第一阈值，且小于或等于所述第二阈值时，所述第二非线性亮度信号的亮度值基于以所述第一非线性亮度信号的亮度值 为自变量的拟合曲线获得；
    当所述第一非线性亮度信号的亮度值大于所述第二阈值时，所述第二非线性亮度信号的亮度值等于所述显示设备对应的最大非线性显示亮度值。
  7. 根据权利要求6所述的方法，所述拟合曲线通过对所述第一阈值和所述第二阈值进行埃尔米特Hermite插值得到。
  8. 根据权利要求6或7任一项所述的方法，其特征在于，所述对所述第一非线性亮度信号进行分段亮度映射，以得到第二非线性亮度信号，包括：
    使用以下分段函数对所述第一非线性亮度信号进行所述亮度映射：
    Figure PCTCN2019090566-appb-100001
    其中，hmt(x)＝y 0×α 0(x)+y 1×α 1(x)+y′ 0β 0(x)+y′ 1β 1(x)
    Figure PCTCN2019090566-appb-100002
    e为所述第一非线性亮度信号，f tm(e)为所述第二非线性亮度信号，KP1为所述第一阈值，KP2为所述第二阈值，maxDL为所述显示设备的最大非线性显示亮度值，maxSL为最大非线性源亮度值，x 0＝KP1，x 1＝maxSL，y 0＝KP1，y 1＝maxDL，y 0′＝1，y 1′＝0。
  9. 根据权利要求6至8任一项所述的方法，其特征在于，所述确定第一阈值和第二阈值，包括：
    根据所述第一非线性亮度信号的显示亮度范围和所述显示设备的显示亮度范围的关系确定所述第一阈值；
    将所述第一非线性亮度信号的最大亮度值作为所述第二阈值。
  10. 根据权利要求1至5任一项所述的方法，其特征在于，所述对所述第一非线性亮度信号进行分段亮度映射，以得到第二非线性亮度信号，包括：
    基于预设的所述第一非线性亮度信号和所述第二非线性亮度信号的亮度值的映射关系，确定与所述第一非线性亮度信号的亮度值相对应的第二非线性亮度信号的亮度值。
  11. 根据权利要求1至10任一项所述的方法，其特征在于，所述将所述第一线性亮度信号转换为第一非线性亮度信号，包括：
    根据PQ电光转移逆函数，将所述第一线性亮度信号转换为所述第一非线性亮度信号；
    对应的，所述将所述第二非线性亮度信号转换为第二线性亮度信号，包括：
    根据PQ电光转移函数，将所述第二非线性亮度信号转换为所述第二线性亮度信号。
  12. 一种视频信号处理的装置，其特征在于，所述装置包括：
    亮度获取单元，用于获得第一线性亮度信号，所述第一线性亮度信号基于待处理视频信号对应的第一线性红绿蓝RGB信号获得；
    第一转换单元，用于将所述第一线性亮度信号转换为第一非线性亮度信号；
    亮度映射单元，用于对所述第一非线性亮度信号进行分段亮度映射，以得到第二非线性亮度信号；
    第二转换单元，用于将所述第二非线性亮度信号转换为第二线性亮度信号；
    增益计算单元，用于计算所述第二线性亮度信号与所述第一线性亮度信号的亮度增益；
    显示信号获取单元，用于基于所述亮度增益和所述第一线性RGB信号的乘积，获得所述待处理视频信号对应的RGB显示信号。
  13. 根据权利要求12所述的装置，其特征在于，所述待处理视频信号为感知量化PQ信号，所述亮度获取单元，具体用于：
    对所述PQ信号进行色彩空间转换，以获得第一非线性RGB信号；
    根据PQ电光转移函数，将所述第一非线性RGB信号转换为所述第一线性RGB信号；
    基于所述第一线性RGB信号的各基色信号进行计算，以获得所述第一线性亮度信号。
  14. 根据权利要求12所述的装置，其特征在于，所述待处理视频信号为混合对数伽玛HLG信号，所述亮度获取单元，具体用于：
    对所述HLG信号进行色彩空间转换，以获得第二非线性RGB信号；
    根据HLG光电转移逆函数，将所述第二非线性RGB信号转换为第二线性RGB信号；
    基于所述第二线性RGB信号的各基色信号进行计算，以获得第三线性亮度信号；
    对所述第三线性亮度信号进行亮度信号类型转换，以获得所述第一线性亮度信号。
  15. 根据权利要求12至14任一项所述的装置，其特征在于，所述装置还包括：色彩空间转换单元，用于：
    对所述RGB显示信号进行色彩空间转换，以获得目标显示信号，其中，所述目标显示信号的色彩格式和显示设备对应的色彩格式相同。
  16. 根据权利要求15所述的装置，其特征在于，所述装置还包括：补偿单元，用于：
    对所述RGB显示信号的各基色值叠加黑位电平提升BlackLevelLift，以获得处理后的RGB显示信号，所述BlackLevelLift为所述显示设备的显示亮度的最小值；
    对应的，所述色彩空间转换单元，具体用于：
    对所述处理后的RGB显示信号进行色彩空间转换。
  17. 根据权利要求12至16任一项所述的装置，其特征在于，所述亮度映射单元，具体用于：
    确定第一阈值和第二阈值，所述第一阈值小于所述第二阈值；
    当所述第一非线性亮度信号的亮度值小于或等于所述第一阈值时，所述第二非线性亮度信号的亮度值等于所述第一非线性亮度信号的亮度值；
    当所述第一非线性亮度信号的亮度值大于所述第一阈值，且小于或等于所述第二阈值时，所述第二非线性亮度信号的亮度值基于以所述第一非线性亮度信号的亮度值为自变量的拟合曲线获得；
    当所述第一非线性亮度信号的亮度值大于所述第二阈值时，所述第二非线性亮度信号的亮度值等于所述显示设备对应的最大非线性显示亮度值。
  18. 根据权利要求17所述的装置，其特征在于，所述拟合曲线通过对所述第一阈值和所述第二阈值进行埃尔米特Hermite插值得到。
  19. 根据权利要求17或18所述的装置，其特征在于，所述亮度映射单元，具体用于：
    使用以下分段函数对所述第一非线性亮度信号进行所述亮度映射：
    Figure PCTCN2019090566-appb-100003
    其中，hmt(x)＝y 0×α 0(x)+y 1×α 1(x)+y′ 0β 0(x)+y′ 1β 1(x)
    Figure PCTCN2019090566-appb-100004
    e为所述第一非线性亮度信号，f tm(e)为所述第二非线性亮度信号，KP1为所述第一阈值，KP2为所述第二阈值，maxDL为所述显示设备的最大非线性显示亮度值，maxSL为最大非线性源亮度值，x 0＝KP1，x 1＝maxSL，y 0＝KP1，y 1＝maxDL，y 0′＝1，y 1′＝0。
  20. 根据权利要求17至19任一项所述的装置，其特征在于，所述亮度映射单元，具体用于：
    根据所述第一非线性亮度信号的显示亮度范围和所述显示设备的显示亮度范围的关系确定所述第一阈值；
    将所述第一非线性亮度信号的最大亮度值作为所述第二阈值。
  21. 根据权利要求12至16任一项所述的装置，其特征在于，所述亮度映射单元，具体用于：
    基于预设的所述第一非线性亮度信号和所述第二非线性亮度信号的亮度值的映射关系，确定与所述第一非线性亮度信号的亮度值相对应的第二非线性亮度信号的亮度值。
  22. 根据权利要求12至21任一项所述的装置，其特征在于，所述第一转换单元， 具体用于：
    根据PQ电光转移逆函数，将所述第一线性亮度信号转换为所述第一非线性亮度信号；
    对应的，所述第二转换单元，具体用于：
    根据PQ电光转移函数，将所述第二非线性亮度信号转换为所述第二线性亮度信号。
  23. 一种视频信号处理的装置，其特征在于，所述装置包括：处理器和存储器；
    所述处理器，用于调用所述存储器中的软件指令，以执行如权利要求1至11任一项所述的方法。
  24. 一种计算机可读存储介质，所述计算机可读存储介质中存储有指令，当其在计算机或处理器上运行时，使得所述计算机或处理器执行如权利要求1至11任一项所述的方法。
  25. 一种包含指令的计算机程序产品，当其在计算机或处理器上运行时，使得所述计算机或处理器执行如权利要求1至11任一项所述的方法。
    。
PCT/CN2019/090566 2018-07-05 2019-06-10 一种视频信号处理的方法及装置 WO2020007165A1 (zh)

Priority Applications (5)

Application Number Priority Date Filing Date Title
KR1020217000634A KR102511572B1 (ko) 2018-07-05 2019-06-10 비디오 신호 처리 방법 및 장치
JP2020568530A JP7115776B2 (ja) 2018-07-05 2019-06-10 ビデオ信号の処理方法及び装置
CN201980025155.1A CN112272947B (zh) 2018-07-05 2019-06-10 一种视频信号处理的方法、装置及计算机可读存储介质
EP19830987.4A EP3796648A4 (en) 2018-07-05 2019-06-10 VIDEO SIGNAL PROCESSING METHOD AND DEVICE
US17/135,568 US11317071B2 (en) 2018-07-05 2020-12-28 Video signal processing method and apparatus

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN201810732833.5 2018-07-05
CN201810732833 2018-07-05
CN201810797968.X 2018-07-19
CN201810797968.XA CN108900823B (zh) 2018-07-05 2018-07-19 一种视频信号处理的方法及装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/135,568 Continuation US11317071B2 (en) 2018-07-05 2020-12-28 Video signal processing method and apparatus

Publications (1)

Publication Number Publication Date
WO2020007165A1 true WO2020007165A1 (zh) 2020-01-09

Family

ID=64351196

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/090566 WO2020007165A1 (zh) 2018-07-05 2019-06-10 一种视频信号处理的方法及装置

Country Status (6)

Country Link
US (1) US11317071B2 (zh)
EP (1) EP3796648A4 (zh)
JP (1) JP7115776B2 (zh)
KR (1) KR102511572B1 (zh)
CN (2) CN108900823B (zh)
WO (1) WO2020007165A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4053834A1 (en) * 2021-03-05 2022-09-07 Canon Kabushiki Kaisha Image processing apparatus, image capture apparatus, control method, and program
CN117440118A (zh) * 2023-10-23 2024-01-23 镕铭微电子(上海)有限公司 图像处理方法、装置、电子设备及存储介质
CN117440118B (zh) * 2023-10-23 2024-06-11 镕铭微电子(上海)有限公司 图像处理方法、装置、电子设备及存储介质

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108900823B (zh) * 2018-07-05 2019-07-12 华为技术有限公司 一种视频信号处理的方法及装置
CN109616037A (zh) * 2018-12-18 2019-04-12 Oppo广东移动通信有限公司 显示屏的显示校准方法、电子装置及计算机可读存储介质
US11405582B2 (en) * 2019-06-28 2022-08-02 Meta Platforms, Inc. Preprocessing of high-dynamic-range video using a hybrid lookup table scheme
CN110691239B (zh) * 2019-12-11 2020-05-19 武汉精立电子技术有限公司 一种hdr显示模组检测方法、装置及电子设备
US11688361B2 (en) * 2020-06-19 2023-06-27 Apple Inc. Quantifying performance of a display device
CN112750086A (zh) * 2020-08-31 2021-05-04 腾讯科技（深圳）有限公司 一种图像处理方法、装置、电子设备及存储介质
US11508331B1 (en) * 2021-04-28 2022-11-22 Novatek Microelectronics Corp. Image uniformity compensation device
CN113676773B (zh) * 2021-08-16 2023-11-14 广州虎牙信息科技有限公司 一种视频播放方法、系统、装置、计算机设备和存储介质
TWI788983B (zh) * 2021-08-30 2023-01-01 瑞昱半導體股份有限公司 視訊訊號處理裝置與方法
US11769464B2 (en) * 2021-09-02 2023-09-26 Arm Limited Image processing
WO2023064426A1 (en) * 2021-10-13 2023-04-20 Freedom Scientific, Inc. Apparatus and method for reducing photophobia in electronic screens
CN114466147B (zh) * 2021-12-23 2024-03-15 阿里巴巴（中国）有限公司 一种视频亮度的调整方法、装置、电子设备以及存储介质
CN115564659B (zh) * 2022-02-28 2024-04-05 荣耀终端有限公司 视频处理方法和装置
CN114898722A (zh) * 2022-05-25 2022-08-12 深圳创维－Rgb电子有限公司 画面显示方法、装置、设备及计算机可读存储介质
CN115063325A (zh) * 2022-08-17 2022-09-16 中央广播电视总台 一种视频信号处理方法、装置、计算机设备和存储介质
CN115762404B (zh) * 2023-01-10 2023-04-07 禹创半导体(深圳)有限公司 Oled调光的dbv重新映射方法、装置和oled显示设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090310015A1 (en) * 2008-06-16 2009-12-17 El-Mahdy Ahmed Hazem Mohamed R Transformation of a video image from a high dynamic range image to a low dynamic range image
CN105516674A (zh) * 2015-12-24 2016-04-20 潮州响石数码技术有限公司 一种具有hdr显示功能的监视设备
CN106780402A (zh) * 2017-01-16 2017-05-31 聚龙智瞳科技有限公司 基于Bayer格式的图像动态范围扩展方法及装置
CN106878694A (zh) * 2015-12-10 2017-06-20 瑞昱半导体股份有限公司 高动态范围信号处理系统与方法
CN107046628A (zh) * 2017-03-15 2017-08-15 四川长虹电器股份有限公司 一种低亮度8k电视机显示hdr效果的方法
CN108900823A (zh) * 2018-07-05 2018-11-27 华为技术有限公司 一种视频信号处理的方法及装置

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI466093B (zh) * 2007-06-26 2014-12-21 Apple Inc 用於視訊播放的管理技術
US8285059B2 (en) * 2008-05-20 2012-10-09 Xerox Corporation Method for automatic enhancement of images containing snow
JP5275122B2 (ja) 2008-05-30 2013-08-28 パナソニック株式会社 ダイナミックレンジ圧縮装置、ダイナミックレンジ圧縮方法、プログラム、集積回路および撮像装置
KR102060604B1 (ko) * 2013-02-28 2019-12-31 삼성디스플레이 주식회사 휘도 조절부, 이를 포함하는 표시 장치 및 이를 이용한 휘도 조절 방법
CN110460744B (zh) 2014-06-10 2022-03-08 松下知识产权经营株式会社 亮度变换装置和亮度变换方法
JP2016058848A (ja) 2014-09-08 2016-04-21 ソニー株式会社 画像処理装置及び画像処理方法
US20160309154A1 (en) 2015-04-17 2016-10-20 Qualcomm Incorporated Dynamic range adjustment for high dynamic range and wide color gamut video coding
GB2539917B (en) * 2015-06-30 2021-04-07 British Broadcasting Corp Method and apparatus for conversion of HDR signals
BR112018000910A2 (pt) 2015-07-16 2018-09-11 Thomson Licensing método e dispositivo para mapear por tons uma imagem utilizando uma função de ajuste de tom paramétrica
CN106878964B (zh) * 2015-12-14 2020-07-28 中国电信股份有限公司江苏客户运营服务中心 基于短信信道的鉴权系统和方法
EP3235236B1 (en) 2015-12-17 2018-06-20 Koninklijke Philips N.V. Dynamic range coding for images and video
EP3220349A1 (en) * 2016-03-16 2017-09-20 Thomson Licensing Methods, apparatus, and systems for extended high dynamic range ("hdr") hdr to hdr tone mapping
JP7054851B2 (ja) 2016-09-09 2022-04-15 パナソニックＩｐマネジメント株式会社 表示装置および信号処理方法
KR102554379B1 (ko) 2016-10-31 2023-07-11 엘지디스플레이 주식회사 하이 다이나믹 레인지 영상 처리 방법 및 영상 처리 모듈과 그를 이용한 표시 장치
PL3389278T3 (pl) 2016-11-22 2023-03-13 Saturn Licensing, Llc Urządzenie nadawcze, sposób nadawania, urządzenie odbiorcze i sposób odbioru
CN108090879B (zh) * 2017-12-12 2020-11-10 上海顺久电子科技有限公司 一种对输入的高动态范围图像进行处理的方法和显示设备
CN110691277B (zh) * 2018-07-05 2024-03-05 华为技术有限公司 视频信号的处理方法及装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090310015A1 (en) * 2008-06-16 2009-12-17 El-Mahdy Ahmed Hazem Mohamed R Transformation of a video image from a high dynamic range image to a low dynamic range image
CN106878694A (zh) * 2015-12-10 2017-06-20 瑞昱半导体股份有限公司 高动态范围信号处理系统与方法
CN105516674A (zh) * 2015-12-24 2016-04-20 潮州响石数码技术有限公司 一种具有hdr显示功能的监视设备
CN106780402A (zh) * 2017-01-16 2017-05-31 聚龙智瞳科技有限公司 基于Bayer格式的图像动态范围扩展方法及装置
CN107046628A (zh) * 2017-03-15 2017-08-15 四川长虹电器股份有限公司 一种低亮度8k电视机显示hdr效果的方法
CN108900823A (zh) * 2018-07-05 2018-11-27 华为技术有限公司 一种视频信号处理的方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3796648A4

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4053834A1 (en) * 2021-03-05 2022-09-07 Canon Kabushiki Kaisha Image processing apparatus, image capture apparatus, control method, and program
US11770511B2 (en) 2021-03-05 2023-09-26 Canon Kabushiki Kaisha Image processing apparatus, image capture apparatus, control method, and computer-readable storage medium
CN117440118A (zh) * 2023-10-23 2024-01-23 镕铭微电子(上海)有限公司 图像处理方法、装置、电子设备及存储介质
CN117440118B (zh) * 2023-10-23 2024-06-11 镕铭微电子(上海)有限公司 图像处理方法、装置、电子设备及存储介质

Also Published As

Publication number Publication date
KR102511572B1 (ko) 2023-03-17
EP3796648A1 (en) 2021-03-24
JP7115776B2 (ja) 2022-08-09
CN108900823A (zh) 2018-11-27
EP3796648A4 (en) 2021-06-02
JP2021529339A (ja) 2021-10-28
KR20210019076A (ko) 2021-02-19
CN108900823B (zh) 2019-07-12
US11317071B2 (en) 2022-04-26
CN112272947A (zh) 2021-01-26
US20210152801A1 (en) 2021-05-20
CN112272947B (zh) 2022-05-13

Similar Documents

Publication Publication Date Title
WO2020007165A1 (zh) 一种视频信号处理的方法及装置
US11183143B2 (en) Transitioning between video priority and graphics priority
JP6234920B2 (ja) ハイダイナミックレンジ画像信号生成及び処理
US11032579B2 (en) Method and a device for encoding a high dynamic range picture, corresponding decoding method and decoding device
CN107211130B (zh) 对彩色画面进行编码和解码的方法和装置
WO2020007167A1 (zh) 视频信号的处理方法及装置
KR102509504B1 (ko) 코딩 및 디코딩 방법과 대응하는 디바이스
TW201702988A (zh) 用於高動態範圍色調映射運算子之方法，裝置及系統
KR102385726B1 (ko) 하이 다이내믹 레인지 화상을 코딩된 비트스트림으로 인코딩/디코딩하기 위한 방법 및 장치
WO2021073304A1 (zh) 一种图像处理的方法及装置
CN110691227A (zh) 一种视频信号处理方法及装置
CN115052137A (zh) 一种视频信号处理方法及装置
KR20160058153A (ko) 압축된 및 압축해제된 디지털 이미지들 및 비디오의 디스플레이에서 가시적 결함들을 감소시키기 위한 시스템 및 방법
CN117319620B (zh) 一种hdr预览级现场实时调色方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19830987

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2020568530

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2019830987

Country of ref document: EP

Effective date: 20201216

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 20217000634

Country of ref document: KR

Kind code of ref document: A