WO2019080543A1 - 高动态范围视频拍摄方法及拍摄装置 - Google Patents

高动态范围视频拍摄方法及拍摄装置

Info

Publication number
WO2019080543A1
WO2019080543A1 PCT/CN2018/095274 CN2018095274W WO2019080543A1 WO 2019080543 A1 WO2019080543 A1 WO 2019080543A1 CN 2018095274 W CN2018095274 W CN 2018095274W WO 2019080543 A1 WO2019080543 A1 WO 2019080543A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
dynamic range
high dynamic
pixel block
panoramic
Prior art date
Application number
PCT/CN2018/095274
Other languages
English (en)
French (fr)
Inventor
陈丹
邵志兢
Original Assignee
深圳看到科技有限公司
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 深圳看到科技有限公司 filed Critical 深圳看到科技有限公司
Priority to US16/758,405 priority Critical patent/US11070741B2/en
Publication of WO2019080543A1 publication Critical patent/WO2019080543A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
    • H04N25/50—Control of the SSIS exposure
    • H04N25/57—Control of the dynamic range
    • H04N25/58—Control of the dynamic range involving two or more exposures
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules
    • H04N23/698—Control of cameras or camera modules for achieving an enlarged field of view, e.g. panoramic image capture
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformations in the plane of the image
    • G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
    • G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/70—Circuitry for compensating brightness variation in the scene
    • H04N23/73—Circuitry for compensating brightness variation in the scene by influencing the exposure time
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/70—Circuitry for compensating brightness variation in the scene
    • H04N23/741—Circuitry for compensating brightness variation in the scene by increasing the dynamic range of the image compared to the dynamic range of the electronic image sensors
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/80—Camera processing pipelines; Components thereof

Definitions

  • the present invention relates to the field of image processing technologies, and in particular, to a high dynamic range video capturing method and a photographing apparatus.
  • HDR High Dynamic Range
  • the HDR image can make full use of the high sensitivity range of the human eye to display a high-definition image. Since the natural light that the human eye can capture has a high brightness range, it is generally necessary to use an integer of 10 bits to 12 bits to represent the sensitized range of the human eye; and the light sensitivity range of the camera light sensor is often small, generally 8 Bit to indicate the light sensitivity range of the light sensor.
  • the above method of capturing a plurality of camera images of different brightness ranges requires continuously shooting a plurality of images of different exposure settings in a short time, and some of the images are underexposed to better capture the details of the highlights; some images are overexposed. In order to better capture the details of the shadows.
  • the above continuous shooting requires the camera to remain stationary, and the object cannot be quickly moved within the scene to ensure the accuracy of the synthesized HDR image.
  • Embodiments of the present invention provide a high dynamic range video capturing method and a photographing apparatus capable of capturing HDR video on a fast moving object; to solve the existing high dynamic range video shooting method and high speed movement occurring in the HDR video synthesized by the photographing apparatus Technical problem of inaccurate location of the object.
  • Embodiments of the present invention provide a high dynamic range video capture method for multi-lens panoramic camera for capturing high dynamic range video, the multi-lens panoramic camera having a plurality of shooting directions of different shooting directions, and adjacent shooting lenses There are overlapping regions between the captured images, wherein the high dynamic range video capturing method includes:
  • the step of determining the image synthesis weight of the corresponding image pixel block according to the pixel color gray scale change amount, the pixel color gray scale difference amount, and the pixel color gray scale offset amount of the image pixel block of the captured image includes:
  • L1 is the pixel color grayscale variation of the image pixel block
  • L2 is the pixel color grayscale difference of the image pixel block
  • L3 is the image color gray
  • a is the pixel color gray scale change amount weight coefficient
  • b is the pixel color gray scale difference amount weight coefficient
  • c is the image color gray scale offset weight coefficient
  • the step of performing brightness continuous processing on the high dynamic range panoramic captured image includes:
  • N is the number of shot lenses
  • I represents the image pixel block in the captured image
  • T represents the image pixel block in the high dynamic range panoramic captured image
  • O represents the high dynamic range panoramic captured image after the brightness continuous processing Image pixel block
  • v k (I n ) is the pixel color grayscale value of the kth image pixel block of the nth shot
  • v k (T) is a pixel color grayscale value of the kth image pixel block of the high dynamic range panoramic image
  • v k (0) is a kth image pixel block of the high dynamic range panoramic captured image after the brightness continuous processing Pixel color grayscale value.
  • Embodiments of the present invention provide a high dynamic range video capture method for multi-lens panoramic camera for capturing high dynamic range video, the multi-lens panoramic camera having a plurality of shooting directions of different shooting directions, and adjacent shooting lenses There are overlapping regions between the captured images, wherein the high dynamic range video capturing method includes:
  • All of the high dynamic range panoramic captured images are stitched together to form a high dynamic range panoramic video.
  • the exposure intensity is set to a high exposure intensity and a low exposure intensity, and when the shooting lens is a high exposure intensity, adjacent shooting lenses are low exposure intensity; When the photographing lens has a low exposure intensity, adjacent photographing lenses are high in exposure intensity.
  • the pixel color gray scale change amount, the pixel color gray scale difference amount, and the pixel color gray scale offset amount of the image pixel block of the captured image are determined according to the corresponding
  • the steps of image composition weighting of image pixel blocks include:
  • L1 is the pixel color grayscale variation of the image pixel block
  • L2 is the pixel color grayscale difference of the image pixel block
  • L3 is the image color gray
  • a is the pixel color gray scale change amount weight coefficient
  • b is the pixel color gray scale difference amount weight coefficient
  • c is the image color gray scale offset weight coefficient.
  • the step of forming the high dynamic range panoramic captured image further includes:
  • the steps of splicing all of the high dynamic range panoramic captured images to form a high dynamic range panoramic video are:
  • All of the processed high dynamic range panoramic captured images are stitched together to form a high dynamic range panoramic video.
  • the step of performing brightness continuous processing on the high dynamic range panoramic captured image includes:
  • N is the number of shot lenses
  • I represents the image pixel block in the captured image
  • T represents the image pixel block in the high dynamic range panoramic captured image
  • O represents the high dynamic range panoramic captured image after the brightness continuous processing Image pixel block
  • v k (I n ) is the pixel color grayscale value of the kth image pixel block of the nth shot
  • v k (T) is a pixel color grayscale value of the kth image pixel block of the high dynamic range panoramic image
  • v k (0) is a kth image pixel block of the high dynamic range panoramic captured image after the brightness continuous processing Pixel color grayscale value.
  • the pixel color gray scale change amount, the pixel color gray scale difference amount, and the pixel color gray scale offset amount of the image pixel block of the captured image are determined according to the corresponding
  • the step of combining image weights of image pixel blocks further includes:
  • the embodiment of the present invention further provides a high dynamic range video capturing device, which is disposed in a multi-lens panoramic camera for performing shooting of a high dynamic range video, the multi-lens panoramic camera having a plurality of different a shooting lens of a shooting direction, and an overlapping area between the captured images of adjacent shooting lenses, wherein the high dynamic range video capturing device comprises:
  • a captured image acquisition module configured to acquire, by using the photographing lens of the multi-lens panoramic camera, a corresponding captured image, wherein an exposure intensity of a captured image of an adjacent photographing lens is different;
  • an image synthesis weight determining module configured to determine an image synthesis weight of the corresponding image pixel block according to the pixel color gray scale change amount, the pixel color gray scale difference amount, and the pixel color gray scale offset amount of the image pixel block of the captured image;
  • An image synthesis module for synthesizing all captured images based on image composition weights of image pixel blocks of each of the captured images to form a high dynamic range panoramic captured image
  • An image splicing module is configured to splicing all of the high dynamic range panoramic captured images to form a high dynamic range panoramic video.
  • the image synthesis weight determination module acquires image synthesis weights according to the following formula:
  • L1 is the pixel color grayscale variation of the image pixel block
  • L2 is the pixel color grayscale difference of the image pixel block
  • L3 is the image color gray
  • a is the pixel color gray scale change amount weight coefficient
  • b is the pixel color gray scale difference amount weight coefficient
  • c is the image color gray scale offset weight coefficient.
  • the high dynamic range video capture device further includes:
  • a brightness continuous processing module configured to perform brightness continuous processing on the high dynamic range panoramic captured image to obtain a processed high dynamic range panoramic captured image
  • the image splicing module is configured to splicing all of the processed high dynamic range panoramic captured images to form a high dynamic range panoramic video.
  • the brightness continuous processing module performs brightness continuous processing on the high dynamic range panoramic captured image according to the following formula:
  • N is the number of shot lenses
  • I represents the image pixel block in the captured image
  • T represents the image pixel block in the high dynamic range panoramic captured image
  • O represents the high dynamic range panoramic captured image after the brightness continuous processing Image pixel block
  • v k (I n ) is the pixel color grayscale value of the kth image pixel block of the nth shot
  • v k (T) is a pixel color grayscale value of the kth image pixel block of the high dynamic range panoramic image
  • v k (0) is a kth image pixel block of the high dynamic range panoramic captured image after the brightness continuous processing Pixel color grayscale value.
  • the high dynamic range video photographing method of the present invention and the photographed image of the adjacent photographing lens of the multi-lens panoramic camera corresponding to the photographing apparatus adopt different exposure intensities, thereby A plurality of captured images of different exposure intensities are combined to form a high dynamic range panoramic image, thereby forming a high dynamic range panoramic video. Since each frame corresponds to at least one pair of high dynamic range panoramic images, HDR video can be taken on fast moving objects; the existing high dynamic range video shooting method and high speed moving objects appearing in the HDR video synthesized by the camera are solved. The location is not accurate technical issues.
  • FIG. 1 is a flow chart of a first embodiment of a high dynamic range video capture method of the present invention
  • FIG. 2 is a flow chart of a second embodiment of a high dynamic range video capture method of the present invention.
  • FIG. 3 is a schematic structural view of a first embodiment of a high dynamic range video capture apparatus according to the present invention.
  • FIG. 4 is a schematic structural view of a second embodiment of a high dynamic range video capture device of the present invention.
  • 5A is a video capture flowchart of a specific embodiment of a high dynamic range video capture method and a high dynamic range video capture apparatus according to the present invention
  • 5B is a schematic diagram of a high dynamic range panoramic video of a high dynamic range video capture method and a high dynamic range video capture apparatus according to the present invention
  • FIG. 6A is a captured image of a high exposure intensity taken by the photographing lens B;
  • FIG. 6B is a captured image of low exposure intensity taken by the photographing lens C.
  • the high dynamic range video shooting method and the photographing apparatus of the present invention can be disposed in a multi-lens panoramic camera, so that the multi-lens panoramic camera can perform high dynamic range video shooting.
  • the multi-lens panoramic camera has a plurality of shooting lenses of different shooting directions, and there are overlapping regions between the captured images of adjacent shooting lenses.
  • the exposure images of the adjacent shooting lenses have different exposure intensities, so that a plurality of captured images of different exposure intensities at the same shooting position can be combined into a high dynamic range panoramic captured image to form a high dynamic range panoramic video.
  • the high dynamic range video shooting method and the photographing apparatus of the present invention can take an HDR video to a fast moving object;
  • FIG. 1 is a flowchart of a first embodiment of a high dynamic range video capturing method according to the present invention.
  • the high dynamic range video capture method of the present embodiment can be implemented using the multi-lens panoramic camera described above, so that the multi-lens panoramic camera performs high dynamic range video capture.
  • the multi-lens panoramic camera here has a plurality of shooting directions of different shooting directions, such as having six shooting lenses, each of which can capture an image area of 180 degrees, such as the above six shooting lenses are evenly distributed at 360 degrees, Each of the image pixel blocks in the finally acquired high dynamic range panoramic captured image corresponds to three shooting lenses.
  • Step S101 the high dynamic range video capturing device acquires the corresponding captured image by using the shooting lens of the multi-lens panoramic camera, wherein the exposure images of the adjacent shooting lenses have different exposure intensities;
  • Step S102 the high dynamic range video capturing device determines the image synthesis weight of the corresponding image pixel block according to the pixel color gray scale change amount, the pixel color gray scale difference amount, and the pixel color gray scale offset amount of the image pixel block of the captured image;
  • Step S103 the high dynamic range video capturing device combines all the captured images based on the image combining weights of the image pixel blocks of each captured image to form a high dynamic range panoramic captured image;
  • step S104 the high dynamic range video capture device splices all of the high dynamic range panoramic captured images to form a high dynamic range panoramic video.
  • step S101 the high dynamic range video capturing device acquires the corresponding captured image by using the shooting lens of the multi-lens panoramic camera, that is, each shooting lens acquires a captured image of its corresponding shooting range, because the captured image of the adjacent shooting lens is There is an overlapping area between them, and the exposure images of the adjacent shooting lenses have different exposure intensities.
  • Each of the regions of the finally synthesized panoramic captured image can be synthesized by the captured images of different exposure intensities, thereby realizing a panoramic captured image having a high dynamic range.
  • the exposure intensity here is divided into high exposure intensity and low exposure intensity.
  • the high exposure intensity refers to the operation of the corresponding captured image of the shooting lens for a long time exposure
  • the low exposure intensity here refers to the corresponding shooting lens acquisition.
  • the captured image is subjected to a short-time exposure operation, and the specific exposure time amount can be set according to the user's request, and the exposure time of the high exposure intensity must be greater than the exposure time of the low exposure intensity.
  • the exposure intensity of all the shooting lenses of the multi-lens panoramic camera is arranged at intervals according to the above-mentioned high exposure intensity and low exposure intensity, that is, when a certain shooting lens is set to a high exposure intensity, other shooting lenses adjacent to the shooting lens should be set. For low exposure intensity; when a shooting lens is set to low exposure intensity, other shooting lenses adjacent to the shooting lens should be set to high exposure intensity.
  • the multi-lens panoramic camera can keep the exposure intensity of each shooting lens constant during shooting, because each shooting area has high exposure intensity and low exposure intensity.
  • the shooting lens is used for shooting, so the finer processing of the captured image can be better achieved.
  • the multi-lens panoramic camera can switch the exposure intensity of the shooting lens during shooting. If the frame is set after the interval, the high-exposure shooting lens is switched to low. Exposure intensity, low exposure intensity shooting mastery switch to high exposure intensity.
  • each shooting lens has a high shooting accuracy for a specific area (such as a facing area facing the front), such as a shooting picture with a high exposure intensity and a low exposure intensity shooting picture for a specific area of the same shooting lens, On the basis of ensuring accurate acquisition of fast moving objects, the accuracy of the static highlight detail (low exposure intensity) and the static dark detail (high exposure intensity) of a specific area are further improved.
  • Step S102 the pixel color grayscale change amount, the pixel color grayscale difference amount, and the pixel color grayscale offset amount of the image pixel block of the captured image acquired by the high dynamic range video capturing device.
  • the image pixel block here is one or more pixel units constituting the captured image.
  • the pixel color gray scale change amount of the image pixel block refers to the change amount of the color gray scale between the current image pixel block and the surrounding adjacent image pixel block, and is used to indicate that the image pixel block is in the captured image and the adjacent image.
  • the display difference of the pixel block If the green grayscale of the current image pixel block is 50 and the adjacent image pixel block has a green grayscale of 51, the green grayscale variation of the image pixel block is 1.
  • the pixel color grayscale variation of the image pixel block herein takes into account the amount of change between all the color grayscales of the current image pixel block and the color grayscale corresponding to all surrounding image pixel blocks. The larger the amount of change, the larger the grayscale variation of the pixel color of the corresponding image pixel block.
  • the pixel color grayscale difference amount of the image pixel refers to the difference value between the red grayscale, the blue grayscale, and the green grayscale in the current image pixel block, and is used to indicate the display difference of the image pixel block on a certain color. For example, if the current image pixel block has a green grayscale of 255, a blue grayscale of 50, and a red grayscale of 50, it may indicate that the image pixel block is biased to display green; for example, the current image pixel block has a green grayscale of 100. The blue gray scale is 100, and the red gray scale is 100, indicating that the image pixel block is biased to display white. When the image pixel block has a greater display difference in a certain color, the pixel color grayscale difference amount of the image pixel block is larger.
  • the pixel color grayscale offset of the image pixel refers to the difference between the color grayscale and the central grayscale of the red grayscale, the blue grayscale, and the green grayscale in the current image pixel block, and is used to indicate the display of the pixel block of the image.
  • brightness If the color gray scale is much higher than the center gray scale, the display brightness of the corresponding image pixel block is brighter; if the color gray scale is much lower than the center gray scale, the display brightness of the corresponding image pixel block is dark. The larger the difference between the color gray scale and the center gray scale, the larger the pixel color gray scale offset of the image pixel block.
  • the high dynamic range video capture device determines the image synthesis weight of the corresponding image pixel block according to the pixel color gray scale change amount, the pixel color gray scale difference amount, and the pixel color gray scale offset amount of the image pixel block of the captured image.
  • the image synthesis weight here is the composite weight ratio of the image pixel block of the captured image when synthesizing the high dynamic range panoramic captured image.
  • the image synthesis weight of the image pixel block can be calculated by the following formula:
  • L1 is the pixel color grayscale variation of the image pixel block
  • L2 is the pixel color grayscale difference of the image pixel block
  • L3 is the image color grayscale offset
  • a is the pixel Color gray scale change amount weight coefficient
  • b is the pixel color gray scale difference amount weight coefficient
  • c is the image color gray scale offset weight coefficient.
  • the image composition weight of the image pixel block is larger; when the pixel color gray scale difference amount is larger, the image synthesis weight of the image pixel block is larger; the image color gray scale offset is larger. In hours, the image composition weight of the image pixel block is larger.
  • step S103 the high dynamic range video capturing device acquires the image combining weights of the image pixel blocks of each captured image based on step S102, and combines all the captured images to form a high dynamic range panoramic captured image.
  • each image pixel block of each captured image has an image composition weight
  • each image pixel block in the synthesized high dynamic range panoramic captured image is composed of at least one image pixel block of the captured image. Therefore, the pixel color grayscale value of each image pixel block in the high dynamic range panoramic captured image can be formed by superimposing a plurality of image pixel blocks. Specifically, the pixel color grayscale value of each image pixel block of the high dynamic range panoramic captured image can be calculated by the following formula:
  • N is the number of shots
  • I is the image pixel block in the captured image
  • T is the image pixel block in the high dynamic range panoramic image
  • v k (I n ) is the kth of the nth shot a pixel color grayscale value of the image pixel block
  • vk (T) is a pixel color grayscale value of the kth image pixel block of the high dynamic range panoramic captured image.
  • step S104 the high dynamic range video capturing device splices all the acquired high dynamic range panoramic captured images in chronological order to finally form a high dynamic range panoramic video.
  • the multi-lens panoramic camera is set to switch the shooting intensity of the shooting lens during the shooting process in step S101, the brightness of the picture frame before and after the switching needs to be continuously processed to reduce the brightness jump of the adjacent picture frame. .
  • the captured image of the adjacent shooting lens of the multi-lens panoramic camera corresponding to the high dynamic range video capturing method of the embodiment adopts different exposure intensities, so that a high dynamic range panoramic captured image can be synthesized by capturing images of different exposure intensities, and further Form a high dynamic range panoramic video. Since each frame of the picture corresponds to at least one pair of high dynamic range panoramic captured images, HDR video can be taken on fast moving objects.
  • FIG. 2 is a flowchart of a second embodiment of a high dynamic range video capturing method according to the present invention.
  • the high dynamic range video capture method of the present embodiment can be implemented using the multi-lens panoramic camera described above, so that the multi-lens panoramic camera performs high dynamic range video capture.
  • the high dynamic range video shooting method of this embodiment includes:
  • Step S201 the high dynamic range video capturing device acquires the corresponding captured image by using the shooting lens of the multi-lens panoramic camera, wherein the exposure images of the adjacent shooting lenses have different exposure intensities;
  • Step S202 the high dynamic range video capturing device determines the image synthesis weight of the corresponding image pixel block according to the pixel color gray scale change amount, the pixel color gray scale difference amount, and the pixel color gray scale offset amount of the image pixel block of the captured image;
  • Step S203 the high dynamic range video capturing device normalizes the image combining weights
  • Step S204 the high dynamic range video capturing device combines all the captured images based on the image combining weights of the image pixel blocks of each captured image to form a high dynamic range panoramic captured image;
  • Step S205 the high dynamic range video capturing device performs brightness continuous processing on the high dynamic range panoramic captured image to obtain the processed high dynamic range panoramic captured image;
  • step S206 the high dynamic range video capturing device splices all the processed high dynamic range panoramic captured images to form a high dynamic range panoramic video.
  • step S201 the high dynamic range video capturing device acquires the corresponding captured image by using the shooting lens of the multi-lens panoramic camera, that is, each shooting lens acquires a captured image of its corresponding shooting range, because the captured image of the adjacent shooting lens is There is an overlapping area between them, and the exposure images of the adjacent shooting lenses have different exposure intensities.
  • Each of the regions of the finally synthesized panoramic captured image can be synthesized by the captured images of different exposure intensities, thereby realizing a panoramic captured image having a high dynamic range.
  • step S202 the pixel color gray scale change amount, the pixel color gray scale difference amount, and the pixel color gray scale shift amount of the image pixel block of the captured image acquired by the high dynamic range video capturing device.
  • the high dynamic range video capture device determines the image synthesis weight of the corresponding image pixel block according to the pixel color gray scale change amount, the pixel color gray scale difference amount, and the pixel color gray scale offset amount of the image pixel block of the captured image.
  • the image synthesis weight here is the composite weight ratio of the image pixel block of the captured image when synthesizing the high dynamic range panoramic captured image.
  • the image synthesis weight of the image pixel block can be calculated by the following formula:
  • L1 is the pixel color grayscale variation of the image pixel block
  • L2 is the pixel color grayscale difference of the image pixel block
  • L3 is the image color grayscale offset
  • a is the pixel Color gray scale change amount weight coefficient
  • b is the pixel color gray scale difference amount weight coefficient
  • c is the image color gray scale offset weight coefficient.
  • the image composition weight of the image pixel block is larger; when the pixel color gray scale difference amount is larger, the image synthesis weight of the image pixel block is larger; the image color gray scale offset is larger. In hours, the image composition weight of the image pixel block is larger.
  • step S203 the high dynamic range video capturing device normalizes the image combining weights acquired in step S202 to subsequently perform the combining operation of the high dynamic range panoramic captured images.
  • step S204 the high dynamic range video capturing device acquires the image combining weights of the image pixel blocks of each captured image based on step S203, and combines all the captured images to form a high dynamic range panoramic captured image.
  • each image pixel block of each captured image has an image composition weight
  • each image pixel block in the synthesized high dynamic range panoramic captured image is composed of at least one image pixel block of the captured image. Therefore, the pixel color grayscale value of each image pixel block in the high dynamic range panoramic captured image can be formed by superimposing a plurality of image pixel blocks. Specifically, the pixel color grayscale value of each image pixel block of the high dynamic range panoramic captured image can be calculated by the following formula:
  • N is the number of shots
  • I is the image pixel block in the captured image
  • T is the image pixel block in the high dynamic range panoramic image
  • v k (I n ) is the kth of the nth shot a pixel color grayscale value of the image pixel block
  • vk (T) is a pixel color grayscale value of the kth image pixel block of the high dynamic range panoramic captured image.
  • step S205 since the image pixel blocks of the high dynamic range panoramic captured image are synthesized by the image pixel blocks in different captured images, the generated high dynamic range panoramic captured image may have a brightness jump of the adjacent image pixel block. change. Therefore, the high dynamic range video capturing apparatus here needs to perform the brightness continuous processing on the high dynamic range panoramic captured image acquired in step S204 to obtain the processed high dynamic range panoramic captured image.
  • the brightness of the high dynamic range panoramic image can be continuously processed by the following formula:
  • N is the number of shot lenses
  • I represents the image pixel block in the captured image
  • T represents the image pixel block in the high dynamic range panoramic captured image
  • O represents the high dynamic range panoramic captured image after the brightness continuous processing Image pixel block
  • v k (I n ) is the pixel color grayscale value of the kth image pixel block of the nth shot
  • v k (T) is a pixel color grayscale value of the kth image pixel block of the high dynamic range panoramic image
  • v k (0) is a kth image pixel block of the high dynamic range panoramic captured image after the brightness continuous processing Pixel color grayscale value.
  • the luminance difference of the high dynamic range panoramic captured image of the final output luminance continuous processing in the overlapping region of each captured image is continuous, thereby avoiding the problem of luminance jump of adjacent image pixel blocks.
  • step S206 the high dynamic range video capture device splices the acquired high dynamic range panoramic captured images after the luminance continuous processing in time series to finally form a high dynamic range panoramic video.
  • the high dynamic range video capturing method of the embodiment further performs brightness continuous processing on the high dynamic range panoramic captured image, thereby further improving the stability of the acquired high dynamic range panoramic captured image, and further Improve the display of the corresponding high dynamic range panoramic video.
  • FIG. 3 is a schematic structural diagram of a first embodiment of a high dynamic range video capture device according to the present invention.
  • the high dynamic range video capture apparatus of the present embodiment can be implemented using the first embodiment of the high dynamic range video capture method described above.
  • the high dynamic range video capture device is set in a multi-lens panoramic camera for multi-lens panoramic camera for high dynamic range video shooting, and the multi-lens panoramic camera has multiple shooting directions for different shooting directions, and adjacent shooting lenses There is an overlap area between the captured images.
  • the high dynamic range video capturing device 30 of the present embodiment includes a captured image acquiring module 31, an image combining weight determining module 32, an image combining module 33, and an image stitching module 34.
  • the captured image acquisition module 31 is configured to acquire a corresponding captured image using the photographing lens of the multi-lens panoramic camera, wherein the exposure intensity of the captured image of the adjacent shooting lens is different;
  • the image combining weight determining module 32 is configured to use the image according to the captured image
  • the pixel color gray scale change amount, the pixel color gray scale difference amount, and the pixel color gray scale shift amount of the pixel block determine the image synthesis weight of the corresponding image pixel block;
  • the image synthesis module 33 is configured to image pixel block based on each captured image
  • the image synthesis weights are combined to form a high dynamic range panoramic captured image;
  • the image stitching module 34 is used to stitch all of the high dynamic range panoramic captured images to form a high dynamic range panoramic video.
  • the captured image acquiring module 31 first acquires a corresponding captured image using the shooting lens of the multi-lens panoramic camera, that is, each shooting lens acquires a captured image of its corresponding shooting range,
  • the captured images of adjacent shooting lenses have overlapping regions, and the exposure images of the adjacent shooting lenses have different exposure intensities.
  • Each of the regions of the finally synthesized panoramic captured image can be synthesized by the captured images of different exposure intensities, thereby realizing a panoramic captured image having a high dynamic range.
  • the exposure intensity here is divided into high exposure intensity and low exposure intensity.
  • the high exposure intensity refers to the operation of the corresponding captured image of the shooting lens for a long time exposure
  • the low exposure intensity here refers to the corresponding shooting lens acquisition.
  • the captured image is subjected to a short-time exposure operation, and the specific exposure time amount can be set according to the user's request, and the exposure time of the high exposure intensity must be greater than the exposure time of the low exposure intensity.
  • the exposure intensity of all the shooting lenses of the multi-lens panoramic camera is arranged at intervals according to the above-mentioned high exposure intensity and low exposure intensity, that is, when a certain shooting lens is set to a high exposure intensity, other shooting lenses adjacent to the shooting lens should be set. For low exposure intensity; when a shooting lens is set to low exposure intensity, other shooting lenses adjacent to the shooting lens should be set to high exposure intensity.
  • the multi-lens panoramic camera can keep the exposure intensity of the shooting lens constant during the shooting process, and can also adjust the exposure intensity of the shooting lens according to the user's request, as long as the exposure intensity of all the shooting lenses of the multi-lens panoramic camera is as high as above.
  • the exposure intensity and the low exposure intensity are sequentially arranged at intervals, and the exposure intensity state of the photographing lens at a certain moment is not specifically limited.
  • the image synthesis weight determination module 32 then acquires the pixel color grayscale variation, the pixel color grayscale difference, and the pixel color grayscale offset of the image pixel block of the captured image.
  • the image pixel block here is one or more pixel units constituting the captured image.
  • the pixel color gray scale change amount of the image pixel block refers to the change amount of the color gray scale between the current image pixel block and the surrounding adjacent image pixel block, and is used to indicate that the image pixel block is in the captured image and the adjacent image.
  • the display difference of the pixel block If the green grayscale of the current image pixel block is 50 and the adjacent image pixel block has a green grayscale of 51, the green grayscale variation of the image pixel block is 1.
  • the pixel color grayscale variation of the image pixel block herein takes into account the amount of change between all the color grayscales of the current image pixel block and the color grayscale corresponding to all surrounding image pixel blocks. The larger the amount of change, the larger the grayscale variation of the pixel color of the corresponding image pixel block.
  • the pixel color grayscale difference amount of the image pixel refers to the difference value between the red grayscale, the blue grayscale, and the green grayscale in the current image pixel block, and is used to indicate the display difference of the image pixel block on a certain color. For example, if the current image pixel block has a green grayscale of 255, a blue grayscale of 50, and a red grayscale of 50, it may indicate that the image pixel block is biased to display green; for example, the current image pixel block has a green grayscale of 100. The blue gray scale is 100, and the red gray scale is 100, indicating that the image pixel block is biased to display white. When the image pixel block has a greater display difference in a certain color, the pixel color grayscale difference amount of the image pixel block is larger.
  • the pixel color grayscale offset of the image pixel refers to the difference between the color grayscale and the central grayscale of the red grayscale, the blue grayscale, and the green grayscale in the current image pixel block, and is used to indicate the display of the pixel block of the image.
  • brightness If the color gray scale is much higher than the center gray scale, the display brightness of the corresponding image pixel block is brighter; if the color gray scale is much lower than the center gray scale, the display brightness of the corresponding image pixel block is dark. The larger the difference between the color gray scale and the center gray scale, the larger the pixel color gray scale offset of the image pixel block.
  • the image synthesis weight determination module 32 determines the image synthesis weight of the corresponding image pixel block according to the pixel color gray scale change amount, the pixel color gray scale difference amount, and the pixel color gray scale offset amount of the image pixel block of the captured image.
  • the image synthesis weight here is the composite weight ratio of the image pixel block of the captured image when synthesizing the high dynamic range panoramic captured image. Specifically, the image synthesis weight of the image pixel block can be calculated by the following formula:
  • L1 is the pixel color grayscale variation of the image pixel block
  • L2 is the pixel color grayscale difference of the image pixel block
  • L3 is the image color grayscale offset
  • a is the pixel Color gray scale change amount weight coefficient
  • b is the pixel color gray scale difference amount weight coefficient
  • c is the image color gray scale offset weight coefficient.
  • the image composition weight of the image pixel block is larger; when the pixel color gray scale difference amount is larger, the image synthesis weight of the image pixel block is larger; the image color gray scale offset is larger. In hours, the image composition weight of the image pixel block is larger.
  • the image synthesis module 33 then acquires the image composition weights of the image pixel blocks of each captured image based on the image synthesis weight determination module 32, and combines all the captured images to form a high dynamic range panoramic captured image.
  • each image pixel block of each captured image has an image composition weight
  • each image pixel block in the synthesized high dynamic range panoramic captured image is composed of at least one image pixel block of the captured image. Therefore, the pixel color grayscale value of each image pixel block in the high dynamic range panoramic captured image can be formed by superimposing a plurality of image pixel blocks. Specifically, the pixel color grayscale value of each image pixel block of the high dynamic range panoramic captured image can be calculated by the following formula:
  • N is the number of shots
  • I is the image pixel block in the captured image
  • T is the image pixel block in the high dynamic range panoramic image
  • v k (I n ) is the kth of the nth shot a pixel color grayscale value of the image pixel block
  • vk (T) is a pixel color grayscale value of the kth image pixel block of the high dynamic range panoramic captured image.
  • the final image splicing module 34 splices all of the acquired high dynamic range panoramic captured images in chronological order to finally form a high dynamic range panoramic video.
  • the captured image of the adjacent shooting lens of the multi-lens panoramic camera corresponding to the high dynamic range video capturing device of the embodiment adopts different exposure intensities, so that a high dynamic range panoramic captured image can be synthesized through a plurality of captured images of different exposure intensities, and further Form a high dynamic range panoramic video. Since each frame of the picture corresponds to at least one pair of high dynamic range panoramic captured images, HDR video can be taken on fast moving objects.
  • FIG. 4 is a schematic structural diagram of a second embodiment of a high dynamic range video capture apparatus according to the present invention.
  • the high dynamic range video capture apparatus of the present embodiment can be implemented using the second embodiment of the high dynamic range video capture method described above.
  • the high dynamic range video capturing device 40 of the present embodiment includes a captured image acquiring module 41, an image combining weight determining module 42, an image combining module 43, a brightness continuous processing module 44, and an image stitching module 45.
  • the captured image acquisition module 41 is configured to acquire a corresponding captured image using a photographing lens of the multi-lens panoramic camera, wherein the exposure intensity of the captured image of the adjacent shooting lens is different; the image combining weight determining module 42 is configured to use the image pixel block of the captured image The pixel color gray scale change amount, the pixel color gray scale difference amount, and the pixel color gray scale offset amount determine the image synthesis weight of the corresponding image pixel block; the image synthesis module 43 is configured to image pixel block based on each of the captured images Image synthesis weights, all captured images are combined to form a high dynamic range panoramic captured image; brightness continuous processing module 44 is used to perform brightness continuous processing on high dynamic range panoramic captured images to obtain processed high dynamics The range merging image is used; the image splicing module 45 is configured to splicing all the processed high dynamic range panoramic captured images to form a high dynamic range panoramic video.
  • the first captured image acquiring module 41 acquires a corresponding captured image using the capturing lens of the multi-lens panoramic camera, that is, each shooting lens acquires a captured image of its corresponding shooting range,
  • the captured images of adjacent shooting lenses have overlapping regions, and the exposure images of the adjacent shooting lenses have different exposure intensities.
  • Each of the regions of the finally synthesized panoramic captured image can be synthesized by the captured images of different exposure intensities, thereby realizing a panoramic captured image having a high dynamic range.
  • the image color grayscale change amount, the pixel color grayscale difference amount, and the pixel color grayscale shift amount of the image pixel block of the captured image acquired by the image synthesis weight determination module 42 is the image color grayscale change amount, the pixel color grayscale difference amount, and the pixel color grayscale shift amount of the image pixel block of the captured image acquired by the image synthesis weight determination module 42.
  • the image synthesis weight determination module 42 determines the image synthesis weight of the corresponding image pixel block according to the pixel color gray scale change amount, the pixel color gray scale difference amount, and the pixel color gray scale offset amount of the image pixel block of the captured image.
  • the image synthesis weight here is the composite weight ratio of the image pixel block of the captured image when synthesizing the high dynamic range panoramic captured image. Specifically, the image synthesis weight of the image pixel block can be calculated by the following formula:
  • L1 is the pixel color grayscale variation of the image pixel block
  • L2 is the pixel color grayscale difference of the image pixel block
  • L3 is the image color grayscale offset
  • a is the pixel Color gray scale change amount weight coefficient
  • b is the pixel color gray scale difference amount weight coefficient
  • c is the image color gray scale offset weight coefficient.
  • the image composition weight of the image pixel block is larger; when the pixel color gray scale difference amount is larger, the image synthesis weight of the image pixel block is larger; the image color gray scale offset is larger. In hours, the image composition weight of the image pixel block is larger.
  • the image synthesis weight determination module 42 then performs a normalization operation on the acquired image synthesis weights to subsequently perform a synthesis operation of the high dynamic range panoramic captured image.
  • the image synthesis module 43 then acquires the image composition weights of the image pixel blocks of each captured image based on the image synthesis weight determination module, and combines all the captured images to form a high dynamic range panoramic captured image.
  • each image pixel block of each captured image has an image composition weight
  • each image pixel block in the synthesized high dynamic range panoramic captured image is composed of at least one image pixel block of the captured image. Therefore, the pixel color grayscale value of each image pixel block in the high dynamic range panoramic captured image can be formed by superimposing a plurality of image pixel blocks. Specifically, the pixel color grayscale value of each image pixel block of the high dynamic range panoramic captured image can be calculated by the following formula:
  • N is the number of shots
  • I is the image pixel block in the captured image
  • T is the image pixel block in the high dynamic range panoramic image
  • v k (I n ) is the kth of the nth shot a pixel color grayscale value of the image pixel block
  • vk (T) is a pixel color grayscale value of the kth image pixel block of the high dynamic range panoramic captured image.
  • the brightness continuous processing module 44 needs to perform brightness continuous processing on the high dynamic range panoramic captured image acquired by the image combining module to obtain the processed high dynamic range panoramic captured image.
  • the brightness of the high dynamic range panoramic image can be continuously processed by the following formula:
  • N is the number of shot lenses
  • I represents the image pixel block in the captured image
  • T represents the image pixel block in the high dynamic range panoramic captured image
  • O represents the high dynamic range panoramic captured image after the brightness continuous processing Image pixel block
  • v k (I n ) is the pixel color grayscale value of the kth image pixel block of the nth shot
  • v k (T) is a pixel color grayscale value of the kth image pixel block of the high dynamic range panoramic image
  • v k (0) is a kth image pixel block of the high dynamic range panoramic captured image after the brightness continuous processing Pixel color grayscale value.
  • the luminance difference of the high dynamic range panoramic captured image after the brightness continuous processing of the brightness continuous processing module 44 is continuous in the overlapping area of each captured image, thereby avoiding the problem of brightness jump of adjacent image pixel blocks. .
  • the final image splicing module 45 splices the acquired high dynamic range panoramic captured images after the brightness is continuously processed in time series to finally form a high dynamic range panoramic video.
  • the high dynamic range video capturing device of the embodiment further performs brightness continuous processing on the high dynamic range panoramic captured image, thereby further improving the stability of the acquired high dynamic range panoramic captured image, and further Improve the display of the corresponding high dynamic range panoramic video.
  • FIG. 5A is a video capture flowchart of a high dynamic range video capture method and a high dynamic range video capture device according to the present invention
  • FIG. 5B is a high dynamic range video capture method of the present invention
  • the high dynamic range video capturing device of the present embodiment is disposed in a multi-lens panoramic camera including a shooting lens A, a shooting lens B, a shooting lens C, a shooting lens D, a shooting lens E, and a shooting lens F, each Each shooting lens can cover a shooting range of 180 degrees, as shown in the shooting range in Figure 5B.
  • the process of the high dynamic range video capture device of the present invention for capturing high dynamic video includes:
  • each of the shooting lenses of the multi-lens panoramic camera acquires a corresponding captured image; the exposure images of the adjacent shooting lenses have different exposure intensities, such as the shooting lens A, the shooting lens C, and the shooting lens E for low-intensity image capturing.
  • the shooting lens B, the shooting lens D, and the shooting lens F perform image shooting with high exposure intensity.
  • FIG. 6A and FIG. 6B FIG. 6A is a captured image of high exposure intensity taken by the photographing lens B, and FIG. 6B is a photographed image of low exposure intensity photographed by the photographing lens C.
  • the area A in FIG. 6A is caused to be missing due to the high exposure intensity, but the content of the A area can be normally displayed due to the lower exposure intensity in FIG. 6B; the image display content is caused by the lower exposure intensity in the area B in FIG. 6B. Missing, but in Fig. 6A, the content of the B region can be normally displayed due to the high exposure intensity.
  • Step S502 the multi-lens panoramic camera acquires a pixel color grayscale change amount, a pixel color grayscale difference amount, and a pixel color grayscale offset amount of the image pixel block of each captured image, so as to find an abnormally displayed image pixel of the captured image.
  • the multi-lens panoramic camera determines image composition weights of the corresponding image pixel blocks according to the pixel color gray scale change amount, the pixel color gray scale difference amount, and the pixel color gray scale shift amount of the image pixel block.
  • the image composition weight of the corresponding image pixel block is high.
  • the image composition weight of the image pixel block of the B area in FIG. 6B is low, and the image synthesis weight of the image pixel block of the A area in FIG. 6B is high.
  • the normalization operation of the image synthesis weight can be selected.
  • step S503 the multi-lens panoramic camera performs a synthesizing operation on all the captured images according to the image synthesis weight of the graphic pixel block of the captured image acquired in step S502 to form a high dynamic range panoramic captured image.
  • the pixel color grayscale value of each image pixel block of the high dynamic range panoramic captured image can be calculated by the following formula:
  • the number of shooting lenses is 6, I represents an image pixel block in the captured image, T represents an image pixel block in the high dynamic range panoramic captured image, and v k (I n ) is the kth of the nth shooting lens.
  • the pixel color grayscale value of the image pixel block; vk (T) is the pixel color grayscale value of the kth image pixel block of the high dynamic range panoramic captured image.
  • Step S504 the multi-lens panoramic camera performs brightness continuous processing on the high dynamic range panoramic captured image by the following formula:
  • N is the number of shot lenses
  • I represents the image pixel block in the captured image
  • T represents the image pixel block in the high dynamic range panoramic captured image
  • O represents the high dynamic range panoramic captured image after the brightness continuous processing Image pixel block
  • v k (I n ) is the pixel color grayscale value of the kth image pixel block of the nth shot
  • v k (T) is a pixel color grayscale value of the kth image pixel block of the high dynamic range panoramic image
  • v k (0) is a kth image pixel block of the high dynamic range panoramic captured image after the brightness continuous processing Pixel color grayscale value.
  • the brightness of the final output is continuously processed, and the high dynamic range panoramic captured image is continuous, and the luminance difference of the captured image in the high dynamic range panoramic image is continuous in the overlapping area of each captured image, thereby avoiding the brightness of the adjacent image pixel block.
  • the problem of jumping is described in detail below.
  • Step S505 the multi-lens panoramic camera splices all the acquired high dynamic range panoramic captured images in chronological order to finally form a high dynamic range panoramic video, as shown in FIG. 5B.
  • the high dynamic range video capturing method of the present invention and the captured image of the adjacent shooting lens of the multi-lens panoramic camera corresponding to the capturing device adopt different exposure intensities, so that a high dynamic range panoramic captured image can be synthesized through a plurality of captured images of different exposure intensities. , which in turn forms a high dynamic range panoramic video. Since each frame corresponds to at least one pair of high dynamic range panoramic images, HDR video can be taken on fast moving objects; the existing high dynamic range video shooting method and high speed moving objects appearing in the HDR video synthesized by the camera are solved. The location is not accurate technical issues.
  • ком ⁇ онент can be, but is not limited to being, a process running on a processor, a processor, an object, an executable application, a thread of execution, a program, and/or a computer.
  • a component can be, but is not limited to being, a process running on a processor, a processor, an object, an executable application, a thread of execution, a program, and/or a computer.
  • an application running on a controller and the controller can be a component.
  • One or more components can reside within a process and/or thread of execution, and a component can be located on a computer and/or distributed between two or more computers.
  • the one or more operations may constitute computer readable instructions stored on one or more computer readable media that, when executed by an electronic device, cause the computing device to perform the operations.
  • the order in which some or all of the operations are described should not be construed as implying that the operations must be sequential. Those skilled in the art will appreciate alternative rankings that have the benefit of this specification. Moreover, it should be understood that not all operations must be present in every embodiment provided herein.
  • the word "preferred” as used herein is intended to serve as an example, instance, or illustration. Any aspect or design described as “preferred” by the text is not necessarily to be construed as being more advantageous than other aspects or designs. Instead, the use of the word “preferred” is intended to present a concept in a specific manner.
  • the term “or” as used in this application is intended to mean an “or” or “an” That is, unless otherwise specified or clear from the context, "X employs A or B” means naturally including any one of the permutations. That is, if X uses A; X uses B; or X uses both A and B, then "X uses A or B" is satisfied in any of the foregoing examples.
  • Each functional unit in the embodiment of the present invention may be integrated into one processing module, or each unit may exist physically separately, or two or more units may be integrated into one module.
  • the above integrated modules can be implemented in the form of hardware or in the form of software functional modules.
  • the integrated modules, if implemented in the form of software functional modules and sold or used as stand-alone products, may also be stored in a computer readable storage medium.
  • the above mentioned storage medium may be a read only memory, a magnetic disk or an optical disk or the like.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Studio Devices (AREA)

Abstract

本发明提供一种高动态范围视频拍摄方法，其包括：使用多镜头全景相机的所述拍摄镜头获取对应的拍摄图像；根据拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重；基于每个拍摄图像的图像像素块的图像合成权重，形成高动态范围全景拍摄图像；将所有的高动态范围全景拍摄图像进行拼接。

Description

高动态范围视频拍摄方法及拍摄装置
本申请要求于2017年10月23日提交中国专利局、申请号为201710994861.X、发明名称为“高动态范围视频拍摄方法及拍摄装置”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
本发明涉及图像处理技术领域，特别是涉及一种高动态范围视频拍摄方法及拍摄装置。
背景技术
HDR(High Dynamic Range)是指高动态范围的图像，HDR图像可以充分利用人眼的高感光范围来展示高精细化的图像。由于人眼能捕捉到的自然界的光线有很高的亮度范围，一般需要使用10位-12位的整数来表示人眼的感光范围；而相机光线感应器的感光范围往往较小，一般使用8位来表示光线感应器的感光范围。
这样导致需要通过将同一位置的多张不同亮度范围的相机图像进行合成才能较好的HDR图像(即有的相机图像捕捉较暗的亮度范围，有的相机图像捕捉较亮的亮度范围)。
上述捕捉多张不同亮度范围的相机图像的方式需要在短时间内连续拍摄多张不同曝光设置的图像，其中有的图像曝光不足，以能够较好的捕捉亮部细节；有的图像曝光过度，以能够较好的捕捉暗部细节。上述连续拍摄的需要相机保持静止不动，且场景内不能够具有快速移动的物体，才能保证合成的HDR图像的准确性。
但是现有的视频拍摄经常会在场景中出现飞鸟、跑动中的人等快速移动的物体，这样导致后期合成的HDR视频会出现物***置不准确的技术问题。
故，有必要提供一种高动态范围视频拍摄方法及拍摄装置，以解决现有技术所存在的问题。
发明内容
本发明实施例提供一种可对快速移动的物体拍摄HDR视频的高动态范围视频拍摄方法及拍摄装置；以解决现有的高动态范围视频拍摄方法及拍摄装置合成的HDR视频中出现的高速移动物体的位置不准确的技术问题。
本发明实施例提供一种高动态范围视频拍摄方法，用于多镜头全景相机进行高动态范围视频的拍摄，所述多镜头全景相机具有多个不同拍摄方向的拍摄镜头，且相邻的拍摄镜头的拍摄图像之间具有重叠区域，其中所述高动态范围视频拍摄方法包括：
使用所述多镜头全景相机的所述拍摄镜头获取对应的拍摄图像，其中相邻的拍摄镜头的拍摄图像的曝光强度不同；
根据所述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重；
基于每个所述拍摄图像的图像像素块的图像合成权重，将所有的拍摄图像进行合成，以形成高动态范围全景拍摄图像；
对所述高动态范围全景拍摄图像进行亮度连续化处理，以得到处理后的高动态范围全景拍摄图像；以 及
将所有的所述处理后的高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频；
所述根据所述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重的步骤包括：
w＝a*L1+b*L2-c*L3；
其中w为所述图像像素块的图像合成权重，L1为所述图像像素块的像素色彩灰阶变化量，L2为所述图像像素块的像素色彩灰阶差异量，L3为所述图像色彩灰阶偏移量，a为像素色彩灰阶变化量权重系数，b为像素色彩灰阶差异量权重系数，c为图像色彩灰阶偏移量权重系数；
其中所述对所述高动态范围全景拍摄图像进行亮度连续化处理的步骤包括：
Figure PCTCN2018095274-appb-000001
Figure PCTCN2018095274-appb-000002
其中N为拍摄镜头的数量，I表示拍摄图像中的图像像素块，T表示所述高动态范围全景拍摄图像中的图像像素块，O表示所述亮度连续化处理后的高动态范围全景拍摄图像中的图像像素块，
Figure PCTCN2018095274-appb-000003
为第n个拍摄镜头的第k个图像像素块的图像合成权重，v k(I n)为第n个拍摄镜头的第k个图像像素块的像素彩色灰阶值；v k(T)为所述高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值；v k(O)为所述亮度连续化处理后的高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值。
本发明实施例提供一种高动态范围视频拍摄方法，用于多镜头全景相机进行高动态范围视频的拍摄，所述多镜头全景相机具有多个不同拍摄方向的拍摄镜头，且相邻的拍摄镜头的拍摄图像之间具有重叠区域，其中所述高动态范围视频拍摄方法包括：
使用所述多镜头全景相机的所述拍摄镜头获取对应的拍摄图像，其中相邻的拍摄镜头的拍摄图像的曝光强度不同；
根据所述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重；
基于每个所述拍摄图像的图像像素块的图像合成权重，将所有的拍摄图像进行合成，以形成高动态范围全景拍摄图像；以及
将所有的所述高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频。
在本发明所述的高动态范围视频拍摄方法中，将曝光强度设定为高曝光强度以及低曝光强度，当所述拍摄镜头为高曝光强度时，相邻的拍摄镜头均为低曝光强度；当所述拍摄镜头为低曝光强度时，相邻的拍摄镜头均为高曝光强度。
在本发明所述的高动态范围视频拍摄方法中，所述根据所述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重的步骤包括：
w＝a*L1+b*L2-c*L3；
其中w为所述图像像素块的图像合成权重，L1为所述图像像素块的像素色彩灰阶变化量，L2为所述图像像素块的像素色彩灰阶差异量，L3为所述图像色彩灰阶偏移量，a为像素色彩灰阶变化量权重系数，b为像素色彩灰阶差异量权重系数，c为图像色彩灰阶偏移量权重系数。
在本发明所述的高动态范围视频拍摄方法中，所述形成所述高动态范围全景拍摄图像的步骤之后还包括：
对所述高动态范围全景拍摄图像进行亮度连续化处理，以得到处理后的高动态范围全景拍摄图像；
所述将所有的所述高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频的步骤为：
将所有的所述处理后的高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频。
在本发明所述的高动态范围视频拍摄方法中，所述对所述高动态范围全景拍摄图像进行亮度连续化处理的步骤包括：
Figure PCTCN2018095274-appb-000004
Figure PCTCN2018095274-appb-000005
其中N为拍摄镜头的数量，I表示拍摄图像中的图像像素块，T表示所述高动态范围全景拍摄图像中的图像像素块，O表示所述亮度连续化处理后的高动态范围全景拍摄图像中的图像像素块，
Figure PCTCN2018095274-appb-000006
为第n个拍摄镜头的第k个图像像素块的图像合成权重，v k(I n)为第n个拍摄镜头的第k个图像像素块的像素彩色灰阶值；v k(T)为所述高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值；v k(O)为所述亮度连续化处理后的高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值。
在本发明所述的高动态范围视频拍摄方法中，所述根据所述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重的步骤还包括：
对所述图像合成权重进行归一化操作。
本发明实施例还提供一种高动态范围视频拍摄装置，其设置在多镜头全景相机中，用于所述多镜头全景相机进行高动态范围视频的拍摄，所述多镜头全景相机具有多个不同拍摄方向的拍摄镜头，且相邻的拍摄镜头的拍摄图像之间具有重叠区域，其中所述高动态范围视频拍摄装置包括：
拍摄图像获取模块，用于使用所述多镜头全景相机的所述拍摄镜头获取对应的拍摄图像，其中相邻的拍摄镜头的拍摄图像的曝光强度不同；
图像合成权重确定模块，用于根据所述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶 差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重；
图像合成模块，用于基于每个所述拍摄图像的图像像素块的图像合成权重，将所有的拍摄图像进行合成，以形成高动态范围全景拍摄图像；以及
图像拼接模块，用于将所有的所述高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频。
在本发明所述的高动态范围视频拍摄装置中，图像合成权重确定模块根据以下公式获取图像合成权重：
w＝a*L1+b*L2-c*L3；
其中w为所述图像像素块的图像合成权重，L1为所述图像像素块的像素色彩灰阶变化量，L2为所述图像像素块的像素色彩灰阶差异量，L3为所述图像色彩灰阶偏移量，a为像素色彩灰阶变化量权重系数，b为像素色彩灰阶差异量权重系数，c为图像色彩灰阶偏移量权重系数。
在本发明所述的高动态范围视频拍摄装置中，所述高动态范围视频拍摄装置还包括：
亮度连续化处理模块，用于对所述高动态范围全景拍摄图像进行亮度连续化处理，以得到处理后的高动态范围全景拍摄图像；
所述图像拼接模块用于将所有的所述处理后的高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频。
在本发明所述的高动态范围视频拍摄装置中，所述亮度连续化处理模块根据以下公式对高动态范围全景拍摄图像进行亮度连续化处理：
Figure PCTCN2018095274-appb-000007
Figure PCTCN2018095274-appb-000008
其中N为拍摄镜头的数量，I表示拍摄图像中的图像像素块，T表示所述高动态范围全景拍摄图像中的图像像素块，O表示所述亮度连续化处理后的高动态范围全景拍摄图像中的图像像素块，
Figure PCTCN2018095274-appb-000009
为第n个拍摄镜头的第k个图像像素块的图像合成权重，v k(I n)为第n个拍摄镜头的第k个图像像素块的像素彩色灰阶值；v k(T)为所述高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值；v k(O)为所述亮度连续化处理后的高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值。
相对现有技术的高动态范围视频拍摄方法及拍摄装置，本发明的高动态范围视频拍摄方法及拍摄装置对应的多镜头全景相机的相邻拍摄镜头的拍摄图像采用不同的曝光强度，从而可通过多张不同曝光强度的拍摄图像合成高动态范围全景拍摄图像，进而形成高动态范围全景视频。由于每帧画面至少对应一副高动态范围全景拍摄图像，因此可对快速移动的物体拍摄HDR视频；解决了现有的高动态范围视频拍摄方法及拍摄装置合成的HDR视频中出现的高速移动物体的位置不准确的技术问题。
附图说明
下面根据附图和实施例对本发明作进一步详细说明。
图1为本发明的高动态范围视频拍摄方法的第一实施例的流程图；
图2为本发明的高动态范围视频拍摄方法的第二实施例的流程图；
图3为本发明的高动态范围视频拍摄装置的第一实施例的结构示意图；
图4为本发明的高动态范围视频拍摄装置的第二实施例的结构示意图；
图5A为本发明的高动态范围视频拍摄方法及高动态范围视频拍摄装置的具体实施例的视频拍摄流程图；
图5B为本发明的高动态范围视频拍摄方法及高动态范围视频拍摄装置的具体实施例的高动态范围全景视频的示意图；
图6A为拍摄镜头B拍摄的高曝光强度的拍摄图像；
图6B为拍摄镜头C拍摄的低曝光强度的拍摄图像。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
本发明的高动态范围视频拍摄方法及拍摄装置可设置在多镜头全景相机中，以便多镜头全景相机进行高动态范围视频的拍摄。该多镜头全景相机具有多个不同拍摄方向的拍摄镜头，且相邻的拍摄镜头的拍摄图像之间具有重叠区域。这里相邻的拍摄镜头的拍摄图像的曝光强度不同，从而同一拍摄位置的多张不同曝光强度的拍摄图像可合成高动态范围全景拍摄图像，进而形成高动态范围全景视频。由于高动态范围全景视频的每个画面帧均对应一高动态范围全景拍摄图像，因此本发明的高动态范围视频拍摄方法及拍摄装置可对快速移动的物体拍摄HDR视频；解决了现有的高动态范围视频拍摄方法及拍摄装置合成的HDR视频中出现的高速移动物体的位置不准确的技术问题。
请参照图1，图1为本发明的高动态范围视频拍摄方法的第一实施例的流程图。本实施例的高动态范围视频拍摄方法可使用上述的多镜头全景相机进行实施，以便该多镜头全景相机进行高动态范围视频的拍摄。这里的多镜头全景相机具有多个不同拍摄方向的拍摄镜头，如具有六个拍摄镜头等，每个拍摄镜头均可拍摄180度的图像区域，如上述六个拍摄镜头以360度均匀分布，则最终获取的高动态范围全景拍摄图像中的每个图像像素块均对应三个拍摄镜头。
本实施例的高动态范围视频拍摄方法包括：
步骤S101，高动态范围视频拍摄装置使用多镜头全景相机的拍摄镜头获取对应的拍摄图像，其中相邻的拍摄镜头的拍摄图像的曝光强度不同；
步骤S102，高动态范围视频拍摄装置根据拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重；
步骤S103，高动态范围视频拍摄装置基于每个拍摄图像的图像像素块的图像合成权重，将所有的拍摄 图像进行合成，以形成高动态范围全景拍摄图像；
步骤S104，高动态范围视频拍摄装置将所有的高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频。
下面详细说明本实施例的高动态范围视频拍摄方法的各步骤的具体流程。
在步骤S101中，高动态范围视频拍摄装置使用多镜头全景相机的拍摄镜头获取对应的拍摄图像，即每个拍摄镜头均获取其对应拍摄范围的拍摄图像，由于相邻的拍摄镜头的拍摄图像之间具有重叠区域，且相邻的拍摄镜头的拍摄图像的曝光强度不同。这样最后合成的全景拍摄图像的每个区域都可由不同曝光强度的拍摄图像合成，从而实现具有高动态范围的全景拍摄图像。
这里的曝光强度分为高曝光强度以及低曝光强度，这里的高曝光强度是指相应的拍摄镜头获取的拍摄图像进行了较长时间曝光的操作；这里的低曝光强度是指相应的拍摄镜头获取的拍摄图像进行了较短时间曝光的操作，具体的曝光时间量可根据用户的要求进行设定，高曝光强度的曝光时间一定大于低曝光强度的曝光时间。多镜头全景相机的所有拍摄镜头的曝光强度按上述高曝光强度和低曝光强度依次间隔排布，即当某个拍摄镜头设置为高曝光强度时，该拍摄镜头相邻的其他拍摄镜头应均设置为低曝光强度；当某个拍摄镜头设置为低曝光强度时，该拍摄镜头相邻的其他拍摄镜头应均设置为高曝光强度。
如用户对高动态范围视频拍摄的精度要求较低，则多镜头全景相机在拍摄过程中可保持每个拍摄镜头的曝光强度一直不变，由于每个拍摄区域都具有高曝光强度和低曝光强度的拍摄镜头进行拍摄，因此可较好的实现拍摄图像的精细化处理。
如用户对高动态范围视频拍摄的精度要求较高，则多镜头全景相机在拍摄过程中可对拍摄镜头的曝光强度进行切换，如间隔设定帧画面后，高曝光强度的拍摄镜头切换为低曝光强度，低曝光强度的拍摄精通切换为高曝光强度。
由于每个拍摄镜头对于特定区域(如正向面对的拍摄区域)的拍摄精度较高，如同一拍摄镜头对特定区域也具有高曝光强度的拍摄画面以及低曝光强度的拍摄画面，可实现在保证对快速移动物体的准确获取的基础上，进一步提高了对特定区域的静态亮部细节(低曝光强度)以及静态暗部细节(高曝光强度)的拍摄精度。
步骤S102，高动态范围视频拍摄装置获取的拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量。
这里的图像像素块为构成该拍摄图像的一个或多个像素单元。其中图像像素块的像素色彩灰阶变化量是指当前图像像素块与周围相邻的图像像素块之间的色彩灰阶的变化量，用于表示该图像像素块在拍摄图像中与相邻图像像素块的显示差异性。如当前图像像素块的绿色灰阶为50，相邻的图像像素块额绿色灰阶为51，则该图像像素块的绿色灰阶变化量为1。当然这里的图像像素块的像素色彩灰阶变化量会考虑当前图像像素块的所有彩色灰阶与周围所有图像像素块对应的彩色灰阶之间的变化量。这个变化量越大，对应的图像像素块的像素色彩灰阶变化量也就越大。
图像像素的像素色彩灰阶差异量是指当前图像像素块中红色灰阶、蓝色灰阶以及绿色灰阶相互之间的差异值，用于表示该图像像素块在某个色彩上的显示差异性，如当前图像像素块的绿色灰阶为255，蓝色灰阶为50，红色灰阶为50，则可表示该图像像素块偏向显示绿色；如当前图像像素块的绿色灰阶为100，蓝色灰阶为100，红色灰阶为100，则表示该图像像素块偏向显示白色。当图像像素块在某个色彩上的显示差异性越大，则该图像像素块的像素色彩灰阶差异量也就越大。
图像像素的像素色彩灰阶偏移量是指当前图像像素块中红色灰阶、蓝色灰阶以及绿色灰阶等彩色灰阶与中心灰阶的差异值，用于表示该图像像素块的显示亮度。如彩色灰阶远高于中心灰阶，则对应的图像像素块的显示亮度较亮；如彩色灰阶远低于中心灰阶，则对应的图像像素块的显示亮度较暗。彩色灰阶与中心灰阶的差异值越大，则该图像像素块的像素色彩灰阶偏移量也就越大。
随后高动态范围视频拍摄装置根据上述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重。这里的图像合成权重为该拍摄图像的图像像素块在合成高动态范围全景拍摄图像时的合成权重比例。具体可通过下式计算图像像素块的图像合成权重：
w＝a*L1+b*L2-c*L3；
其中w为图像像素块的图像合成权重，L1为图像像素块的像素色彩灰阶变化量，L2为图像像素块的像素色彩灰阶差异量，L3为图像色彩灰阶偏移量，a为像素色彩灰阶变化量权重系数，b为像素色彩灰阶差异量权重系数，c为图像色彩灰阶偏移量权重系数。
即当像素色彩灰阶变化量越大时，图像像素块的图像合成权重越大；像素色彩灰阶差异量越大时，图像像素块的图像合成权重越大；图像色彩灰阶偏移量越小时，图像像素块的图像合成权重越大。
步骤S103，高动态范围视频拍摄装置基于步骤S102获取每个拍摄图像的图像像素块的图像合成权重，将所有的拍摄图像进行合成，以形成高动态范围全景拍摄图像。
由于每个拍摄图像的每个图像像素块均有图像合成权重，且合成后的高动态范围全景拍摄图像中的每个图像像素块均由拍摄图像的至少一个图像像素块构成。因此可通过多个图像像素块叠加的方式形成高动态范围全景拍摄图像中的每个图像像素块的像素彩色灰阶值。具体可通过下式计算高动态范围全景拍摄图像的每个图像像素块的像素彩色灰阶值：
Figure PCTCN2018095274-appb-000010
其中N为拍摄镜头的数量，I表示拍摄图像中的图像像素块，T表示所述高动态范围全景拍摄图像中的图像像素块，v k(I n)为第n个拍摄镜头的第k个图像像素块的像素彩色灰阶值；v k(T)为所述高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值。
步骤S104，高动态范围视频拍摄装置将获取的所有高动态范围全景拍摄图像按时间顺序进行拼接，以最终形成高动态范围全景视频。
优选的，如在步骤S101设置了多镜头全景相机在拍摄过程中对拍摄镜头的拍摄强度进行切换，则需要对切换前后的画面帧进行亮度连续化处理，以减轻相邻画面帧的亮度跳变。
这样即完成了本实施例的高动态视频拍摄方法的高动态视频拍摄过程。
本实施例的高动态范围视频拍摄方法对应的多镜头全景相机的相邻拍摄镜头的拍摄图像采用不同的曝光强度，从而可通过多张不同曝光强度的拍摄图像合成高动态范围全景拍摄图像，进而形成高动态范围全景视频。由于每帧画面至少对应一副高动态范围全景拍摄图像，因此可对快速移动的物体拍摄HDR视频。
请参照图2，图2为本发明的高动态范围视频拍摄方法的第二实施例的流程图。本实施例的高动态范围视频拍摄方法可使用上述的多镜头全景相机进行实施，以便该多镜头全景相机进行高动态范围视频的拍摄。本实施例的高动态范围视频拍摄方法包括：
步骤S201，高动态范围视频拍摄装置使用多镜头全景相机的拍摄镜头获取对应的拍摄图像，其中相邻的拍摄镜头的拍摄图像的曝光强度不同；
步骤S202，高动态范围视频拍摄装置根据拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重；
步骤S203，高动态范围视频拍摄装置对图像合成权重进行归一化操作；
步骤S204，高动态范围视频拍摄装置基于每个拍摄图像的图像像素块的图像合成权重，将所有的拍摄图像进行合成，以形成高动态范围全景拍摄图像；
步骤S205，高动态范围视频拍摄装置对高动态范围全景拍摄图像进行亮度连续化处理，以得到处理后的高动态范围全景拍摄图像；
步骤S206，高动态范围视频拍摄装置将所有的处理后的高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频。
下面详细说明本实施例的高动态范围视频拍摄方法的各步骤的具体流程。
在步骤S201中，高动态范围视频拍摄装置使用多镜头全景相机的拍摄镜头获取对应的拍摄图像，即每个拍摄镜头均获取其对应拍摄范围的拍摄图像，由于相邻的拍摄镜头的拍摄图像之间具有重叠区域，且相邻的拍摄镜头的拍摄图像的曝光强度不同。这样最后合成的全景拍摄图像的每个区域都可由不同曝光强度的拍摄图像合成，从而实现具有高动态范围的全景拍摄图像。
在步骤S202中，高动态范围视频拍摄装置获取的拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量。
随后高动态范围视频拍摄装置根据上述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重。这里的图像合成权重为该拍摄图像的图像像素块在合成高动态范围全景拍摄图像时的合成权重比例。具体可通过下式计算图像像素块的图像合成权重：
w＝a*L1+b*L2-c*L3；
其中w为图像像素块的图像合成权重，L1为图像像素块的像素色彩灰阶变化量，L2为图像像素块的像素色彩灰阶差异量，L3为图像色彩灰阶偏移量，a为像素色彩灰阶变化量权重系数，b为像素色彩灰阶差异量权重系数，c为图像色彩灰阶偏移量权重系数。
即当像素色彩灰阶变化量越大时，图像像素块的图像合成权重越大；像素色彩灰阶差异量越大时，图像像素块的图像合成权重越大；图像色彩灰阶偏移量越小时，图像像素块的图像合成权重越大。
在步骤S203中，高动态范围视频拍摄装置对步骤S202获取的图像合成权重进行归一化操作，以便后续进行高动态范围全景拍摄图像的合成操作。
在步骤S204中，高动态范围视频拍摄装置基于步骤S203获取每个拍摄图像的图像像素块的图像合成权重，将所有的拍摄图像进行合成，以形成高动态范围全景拍摄图像。
由于每个拍摄图像的每个图像像素块均有图像合成权重，且合成后的高动态范围全景拍摄图像中的每个图像像素块均由拍摄图像的至少一个图像像素块构成。因此可通过多个图像像素块叠加的方式形成高动态范围全景拍摄图像中的每个图像像素块的像素彩色灰阶值。具体可通过下式计算高动态范围全景拍摄图像的每个图像像素块的像素彩色灰阶值：
Figure PCTCN2018095274-appb-000011
其中N为拍摄镜头的数量，I表示拍摄图像中的图像像素块，T表示所述高动态范围全景拍摄图像中的图像像素块，v k(I n)为第n个拍摄镜头的第k个图像像素块的像素彩色灰阶值；v k(T)为所述高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值。
在步骤S205中，由于高动态范围全景拍摄图像的图像像素块是由不同的拍摄图像中的图像像素块合成，因此生成后的高动态范围全景拍摄图像可能会出现相邻图像像素块的亮度跳变。因此这里高动态范围视频拍摄装置需要对步骤S204获取的高动态范围全景拍摄图像进行亮度连续化处理，以得到处理后的高动态范围的全景拍摄图像。具体可通过以下公式对高动态范围全景拍摄图像进行亮度连续化处理：
Figure PCTCN2018095274-appb-000012
其中N为拍摄镜头的数量，I表示拍摄图像中的图像像素块，T表示所述高动态范围全景拍摄图像中的图像像素块，O表示所述亮度连续化处理后的高动态范围全景拍摄图像中的图像像素块，
Figure PCTCN2018095274-appb-000013
为第n个拍摄镜头的第k个图像像素块的图像合成权重，v k(I n)为第n个拍摄镜头的第k个图像像素块的像素彩色灰阶值；v k(T)为所述高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值；v k(O)为所述亮度连续化处理后的高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值。
最终输出的亮度连续化处理后的高动态范围全景拍摄图像在各个拍摄图像的重叠区域的亮度差都是连续的，从而避免了相邻图像像素块的亮度跳变的问题。
在步骤S206中，高动态范围视频拍摄装置将获取的所有亮度连续化处理后的高动态范围全景拍摄图像按时间顺序进行拼接，以最终形成高动态范围全景视频。
这样即完成了本实施例的高动态视频拍摄方法的高动态视频拍摄过程。
在第一实施例的基础上，本实施例的高动态范围视频拍摄方法对高动态范围全景拍摄图像还进行了亮度连续化处理，从而进一步提高了获取的高动态范围全景拍摄图像稳定性，进而提高了对应的高动态范围全景视频的显示效果。
本发明还提供一种高动态范围视频拍摄装置，请参照图3，图3为本发明的高动态范围视频拍摄装置的第一实施例的结构示意图。本实施例的高动态范围视频拍摄装置可使用上述的高动态范围视频拍摄方法的第一实施例进行实施。该高动态范围视频拍摄装置设置在多镜头全景相机中，用于多镜头全景相机进行高动态范围视频的拍摄，多镜头全景相机具有多个不同拍摄方向的拍摄镜头，且相邻的拍摄镜头的拍摄图像之间具有重叠区域。
本实施例的高动态范围视频拍摄装置30包括拍摄图像获取模块31、图像合成权重确定模块32、图像合成模块33以及图像拼接模块34。拍摄图像获取模块31用于使用多镜头全景相机的所述拍摄镜头获取对应的拍摄图像，其中相邻的拍摄镜头的拍摄图像的曝光强度不同；图像合成权重确定模块32用于根据拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重；图像合成模块33用于基于每个拍摄图像的图像像素块的图像合成权重，将所有的拍摄图像进行合成，以形成高动态范围全景拍摄图像；图像拼接模块34用于将所有的高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频。
本实施例的高动态范围视频拍摄装置30使用时，首先拍摄图像获取模块31使用多镜头全景相机的拍摄镜头获取对应的拍摄图像，即每个拍摄镜头均获取其对应拍摄范围的拍摄图像，由于相邻的拍摄镜头的拍摄图像之间具有重叠区域，且相邻的拍摄镜头的拍摄图像的曝光强度不同。这样最后合成的全景拍摄图像的每个区域都可由不同曝光强度的拍摄图像合成，从而实现具有高动态范围的全景拍摄图像。
这里的曝光强度分为高曝光强度以及低曝光强度，这里的高曝光强度是指相应的拍摄镜头获取的拍摄图像进行了较长时间曝光的操作；这里的低曝光强度是指相应的拍摄镜头获取的拍摄图像进行了较短时间曝光的操作，具体的曝光时间量可根据用户的要求进行设定，高曝光强度的曝光时间一定大于低曝光强度的曝光时间。多镜头全景相机的所有拍摄镜头的曝光强度按上述高曝光强度和低曝光强度依次间隔排布，即当某个拍摄镜头设置为高曝光强度时，该拍摄镜头相邻的其他拍摄镜头应均设置为低曝光强度；当某个拍摄镜头设置为低曝光强度时，该拍摄镜头相邻的其他拍摄镜头应均设置为高曝光强度。
这里多镜头全景相机在拍摄过程中可保持拍摄镜头的曝光强度一直不变，也可根据用户要求对拍摄镜头的曝光强度进行调整，只要保证多镜头全景相机的所有拍摄镜头的曝光强度按上述高曝光强度和低曝光强度依次间隔排布，具体某个时刻的拍摄镜头的曝光强度状态并不进行具体限定。
随后图像合成权重确定模块32获取的拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差 异量以及像素色彩灰阶偏移量。
这里的图像像素块为构成该拍摄图像的一个或多个像素单元。其中图像像素块的像素色彩灰阶变化量是指当前图像像素块与周围相邻的图像像素块之间的色彩灰阶的变化量，用于表示该图像像素块在拍摄图像中与相邻图像像素块的显示差异性。如当前图像像素块的绿色灰阶为50，相邻的图像像素块额绿色灰阶为51，则该图像像素块的绿色灰阶变化量为1。当然这里的图像像素块的像素色彩灰阶变化量会考虑当前图像像素块的所有彩色灰阶与周围所有图像像素块对应的彩色灰阶之间的变化量。这个变化量越大，对应的图像像素块的像素色彩灰阶变化量也就越大。
图像像素的像素色彩灰阶差异量是指当前图像像素块中红色灰阶、蓝色灰阶以及绿色灰阶相互之间的差异值，用于表示该图像像素块在某个色彩上的显示差异性，如当前图像像素块的绿色灰阶为255，蓝色灰阶为50，红色灰阶为50，则可表示该图像像素块偏向显示绿色；如当前图像像素块的绿色灰阶为100，蓝色灰阶为100，红色灰阶为100，则表示该图像像素块偏向显示白色。当图像像素块在某个色彩上的显示差异性越大，则该图像像素块的像素色彩灰阶差异量也就越大。
图像像素的像素色彩灰阶偏移量是指当前图像像素块中红色灰阶、蓝色灰阶以及绿色灰阶等彩色灰阶与中心灰阶的差异值，用于表示该图像像素块的显示亮度。如彩色灰阶远高于中心灰阶，则对应的图像像素块的显示亮度较亮；如彩色灰阶远低于中心灰阶，则对应的图像像素块的显示亮度较暗。彩色灰阶与中心灰阶的差异值越大，则该图像像素块的像素色彩灰阶偏移量也就越大。
然后图像合成权重确定模块32根据上述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重。这里的图像合成权重为该拍摄图像的图像像素块在合成高动态范围全景拍摄图像时的合成权重比例。具体可通过下式计算图像像素块的图像合成权重：
w＝a*L1+b*L2-c*L3；
其中w为图像像素块的图像合成权重，L1为图像像素块的像素色彩灰阶变化量，L2为图像像素块的像素色彩灰阶差异量，L3为图像色彩灰阶偏移量，a为像素色彩灰阶变化量权重系数，b为像素色彩灰阶差异量权重系数，c为图像色彩灰阶偏移量权重系数。
即当像素色彩灰阶变化量越大时，图像像素块的图像合成权重越大；像素色彩灰阶差异量越大时，图像像素块的图像合成权重越大；图像色彩灰阶偏移量越小时，图像像素块的图像合成权重越大。
随后图像合成模块33基于图像合成权重确定模块32获取每个拍摄图像的图像像素块的图像合成权重，将所有的拍摄图像进行合成，以形成高动态范围全景拍摄图像。
由于每个拍摄图像的每个图像像素块均有图像合成权重，且合成后的高动态范围全景拍摄图像中的每个图像像素块均由拍摄图像的至少一个图像像素块构成。因此可通过多个图像像素块叠加的方式形成高动态范围全景拍摄图像中的每个图像像素块的像素彩色灰阶值。具体可通过下式计算高动态范围全景拍摄图像的每个图像像素块的像素彩色灰阶值：
Figure PCTCN2018095274-appb-000014
其中N为拍摄镜头的数量，I表示拍摄图像中的图像像素块，T表示所述高动态范围全景拍摄图像中的图像像素块，v k(I n)为第n个拍摄镜头的第k个图像像素块的像素彩色灰阶值；v k(T)为所述高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值。
最后图像拼接模块34将获取的所有高动态范围全景拍摄图像按时间顺序进行拼接，以最终形成高动态范围全景视频。
这样即完成了本实施例的高动态范围视频拍摄装置30的高动态视频拍摄过程。
本实施例的高动态范围视频拍摄装置对应的多镜头全景相机的相邻拍摄镜头的拍摄图像采用不同的曝光强度，从而可通过多张不同曝光强度的拍摄图像合成高动态范围全景拍摄图像，进而形成高动态范围全景视频。由于每帧画面至少对应一副高动态范围全景拍摄图像，因此可对快速移动的物体拍摄HDR视频。
请参照图4，图4为本发明的高动态范围视频拍摄装置的第二实施例的结构示意图。本实施例的高动态范围视频拍摄装置可使用上述的高动态范围视频拍摄方法的第二实施例进行实施。本实施例的高动态范围视频拍摄装置40包括拍摄图像获取模块41、图像合成权重确定模块42、图像合成模块43、亮度连续化处理模块44以及图像拼接模块45。
拍摄图像获取模块41用于使用多镜头全景相机的拍摄镜头获取对应的拍摄图像，其中相邻的拍摄镜头的拍摄图像的曝光强度不同；图像合成权重确定模块42用于根据拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重；图像合成模块43用于基于每个所述拍摄图像的图像像素块的图像合成权重，将所有的拍摄图像进行合成，以形成高动态范围全景拍摄图像；亮度连续化处理模块44用于对高动态范围全景拍摄图像进行亮度连续化处理，以得到处理后的高动态范围全景拍摄图像；图像拼接模块45用于将所有的处理后的高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频。
本实施例的高动态范围视频拍摄装置40使用时，首先拍摄图像获取模块41使用多镜头全景相机的拍摄镜头获取对应的拍摄图像，即每个拍摄镜头均获取其对应拍摄范围的拍摄图像，由于相邻的拍摄镜头的拍摄图像之间具有重叠区域，且相邻的拍摄镜头的拍摄图像的曝光强度不同。这样最后合成的全景拍摄图像的每个区域都可由不同曝光强度的拍摄图像合成，从而实现具有高动态范围的全景拍摄图像。
随后图像合成权重确定模块42获取的拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量。
随后图像合成权重确定模块42根据上述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重。这里的图像合成权重为该拍摄图像的图像像素块在合成高动态范围全景拍摄图像时的合成权重比例。具体可通过下式计算图像像素块的图像合成权重：
w＝a*L1+b*L2-c*L3；
其中w为图像像素块的图像合成权重，L1为图像像素块的像素色彩灰阶变化量，L2为图像像素块的像素色彩灰阶差异量，L3为图像色彩灰阶偏移量，a为像素色彩灰阶变化量权重系数，b为像素色彩灰阶差异量权重系数，c为图像色彩灰阶偏移量权重系数。
即当像素色彩灰阶变化量越大时，图像像素块的图像合成权重越大；像素色彩灰阶差异量越大时，图像像素块的图像合成权重越大；图像色彩灰阶偏移量越小时，图像像素块的图像合成权重越大。
然后图像合成权重确定模块42对获取的图像合成权重进行归一化操作，以便后续进行高动态范围全景拍摄图像的合成操作。
随后图像合成模块43基于图像合成权重确定模块获取每个拍摄图像的图像像素块的图像合成权重，将所有的拍摄图像进行合成，以形成高动态范围全景拍摄图像。
由于每个拍摄图像的每个图像像素块均有图像合成权重，且合成后的高动态范围全景拍摄图像中的每个图像像素块均由拍摄图像的至少一个图像像素块构成。因此可通过多个图像像素块叠加的方式形成高动态范围全景拍摄图像中的每个图像像素块的像素彩色灰阶值。具体可通过下式计算高动态范围全景拍摄图像的每个图像像素块的像素彩色灰阶值：
Figure PCTCN2018095274-appb-000015
其中N为拍摄镜头的数量，I表示拍摄图像中的图像像素块，T表示所述高动态范围全景拍摄图像中的图像像素块，v k(I n)为第n个拍摄镜头的第k个图像像素块的像素彩色灰阶值；v k(T)为所述高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值。
由于高动态范围全景拍摄图像的图像像素块是由不同的拍摄图像中的图像像素块合成，因此生成后的高动态范围全景拍摄图像可能会出现相邻图像像素块的亮度跳变。因此这里亮度连续化处理模块44需要对图像合成模块获取的高动态范围全景拍摄图像进行亮度连续化处理，以得到处理后的高动态范围的全景拍摄图像。具体可通过以下公式对高动态范围全景拍摄图像进行亮度连续化处理：
Figure PCTCN2018095274-appb-000016
其中N为拍摄镜头的数量，I表示拍摄图像中的图像像素块，T表示所述高动态范围全景拍摄图像中的图像像素块，O表示所述亮度连续化处理后的高动态范围全景拍摄图像中的图像像素块，
Figure PCTCN2018095274-appb-000017
为第n个拍摄镜头的第k个图像像素块的图像合成权重，v k(I n)为第n个拍摄镜头的第k个图像像素块的像素彩色灰阶值；v k(T)为所述高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值；v k(O)为所述亮度连续化处理后的高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值。
最终亮度连续化处理模块44输出的亮度连续化处理后的高动态范围全景拍摄图像在各个拍摄图像的重 叠区域的亮度差都是连续的，从而避免了相邻图像像素块的亮度跳变的问题。
最后图像拼接模块45将获取的所有亮度连续化处理后的高动态范围全景拍摄图像按时间顺序进行拼接，以最终形成高动态范围全景视频。
这样即完成了本实施例的高动态视频拍摄装置40的高动态视频拍摄过程。
在第一实施例的基础上，本实施例的高动态范围视频拍摄装置对高动态范围全景拍摄图像还进行了亮度连续化处理，从而进一步提高了获取的高动态范围全景拍摄图像稳定性，进而提高了对应的高动态范围全景视频的显示效果。
下面通过一具体实施例说明本发明的高动态范围视频拍摄方法及高动态范围视频拍摄装置的工作原理。请参照图5A和图5B，图5A为本发明的高动态范围视频拍摄方法及高动态范围视频拍摄装置的具体实施例的视频拍摄流程图；图5B为本发明的高动态范围视频拍摄方法及高动态范围视频拍摄装置的具体实施例的高动态范围全景视频的示意图。
本具体实施例的高动态范围视频拍摄装置设置在多镜头全景相机中，该多镜头全景相机包括拍摄镜头A、拍摄镜头B、拍摄镜头C、拍摄镜头D、拍摄镜头E以及拍摄镜头F，每个拍摄镜头均可覆盖180度的拍摄范围，具体可参见图5B中的拍摄范围。本发明的高动态范围视频拍摄装置拍摄高动态视频的过程包括：
步骤S501，多镜头全景相机的各个拍摄镜头均获取对应的拍摄图像；相邻的拍摄镜头的拍摄图像的曝光强度不同，如拍摄镜头A、拍摄镜头C以及拍摄镜头E进行低曝光强度的图像拍摄；拍摄镜头B、拍摄镜头D以及拍摄镜头F进行高曝光强度的图像拍摄。具体请参见图6A和图6B，图6A为拍摄镜头B拍摄的高曝光强度的拍摄图像，图6B为拍摄镜头C拍摄的低曝光强度的拍摄图像。图6A中的A区域由于曝光强度较高，导致图像显示内容缺失，但是图6B由于曝光强度较低，A区域内容可以正常显示；图6B中的B区域由于曝光强度较低，导致图像显示内容缺失，但是图6A由于曝光强度较高，B区域内容可以正常显示。
步骤S502，多镜头全景相机获取每个拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量，以便找到拍摄图像的非正常显示的图像像素块以及正常显示的图像像素块。
随后多镜头全景相机根据上述图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重。图6A中的A区域由于像素色彩灰阶偏移量较大、像素色彩灰阶变化量较小且像素色彩灰阶差异量较小，对应的图像像素块的图像合成权重就会较低，而图6A中的B区域由于像素色彩灰阶偏移量较小且像素色彩灰阶差异量较大，对应的图像像素块的图像合成权重就会较高。同理图6B中的B区域的图像像素块的图像合成权重较低，图6B中的A区域的图像像素块的图像合成权重较高。这里可选择对该图像合成权重进行归一化操作。
步骤S503，多镜头全景相机根据步骤S502获取的拍摄图像的图形像素块的图像合成权重，将所有的拍摄图像进行合成操作，以形成高动态范围全景拍摄图像。具体可通过下式计算高动态范围全景拍摄图像的 每个图像像素块的像素彩色灰阶值：
Figure PCTCN2018095274-appb-000018
其中拍摄镜头的数量为6，I表示拍摄图像中的图像像素块，T表示所述高动态范围全景拍摄图像中的图像像素块，v k(I n)为第n个拍摄镜头的第k个图像像素块的像素彩色灰阶值；v k(T)为高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值。
步骤S504，多镜头全景相机通过下式对高动态范围全景拍摄图像进行亮度连续化处理：
Figure PCTCN2018095274-appb-000019
其中N为拍摄镜头的数量，I表示拍摄图像中的图像像素块，T表示所述高动态范围全景拍摄图像中的图像像素块，O表示所述亮度连续化处理后的高动态范围全景拍摄图像中的图像像素块，
Figure PCTCN2018095274-appb-000020
为第n个拍摄镜头的第k个图像像素块的图像合成权重，v k(I n)为第n个拍摄镜头的第k个图像像素块的像素彩色灰阶值；v k(T)为所述高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值；v k(O)为所述亮度连续化处理后的高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值。
从而最终输出的亮度连续化处理后的高动态范围全景拍摄图像，该高动态范围全景内拍摄图像在各个拍摄图像的重叠区域的亮度差都是连续的，从而避免了相邻图像像素块的亮度跳变的问题。
步骤S505，多镜头全景相机将获取的所有高动态范围全景拍摄图像按时间顺序进行拼接，以最终形成高动态范围全景视频，如图5B所示。
这样即完成了本具体实施例的高动态范围视频拍摄方法及高动态范围视频拍摄装置的高动态视频拍摄过程。
本发明的高动态范围视频拍摄方法及拍摄装置对应的多镜头全景相机的相邻拍摄镜头的拍摄图像采用不同的曝光强度，从而可通过多张不同曝光强度的拍摄图像合成高动态范围全景拍摄图像，进而形成高动态范围全景视频。由于每帧画面至少对应一副高动态范围全景拍摄图像，因此可对快速移动的物体拍摄HDR视频；解决了现有的高动态范围视频拍摄方法及拍摄装置合成的HDR视频中出现的高速移动物体的位置不准确的技术问题。
如本申请所使用的术语“组件”、“模块”、“***”、“接口”、“进程”等等一般地旨在指计算机相关实体：硬件、硬件和软件的组合、软件或执行中的软件。例如，组件可以是但不限于是运行在处理器上的进程、处理器、对象、可执行应用、执行的线程、程序和/或计算机。通过图示，运行在控制器上的应用和该控制器二者都可以是组件。一个或多个组件可以有在于执行的进程和/或线程内，并且组件可以位于一个计算机上和/或分布在两个或更多计算机之间。
本文提供了实施例的各种操作。在一个实施例中，所述的一个或多个操作可以构成一个或多个计算机 可读介质上存储的计算机可读指令，其在被电子设备执行时将使得计算设备执行所述操作。描述一些或所有操作的顺序不应当被解释为暗示这些操作必需是顺序相关的。本领域技术人员将理解具有本说明书的益处的可替代的排序。而且，应当理解，不是所有操作必需在本文所提供的每个实施例中存在。
而且，本文所使用的词语“优选的”意指用作实例、示例或例证。奉文描述为“优选的”任意方面或设计不必被解释为比其他方面或设计更有利。相反，词语“优选的”的使用旨在以具体方式提出概念。如本申请中所使用的术语“或”旨在意指包含的“或”而非排除的“或”。即，除非另外指定或从上下文中清楚，“X使用A或B”意指自然包括排列的任意一个。即，如果X使用A；X使用B；或X使用A和B二者，则“X使用A或B”在前述任一示例中得到满足。
而且，尽管已经相对于一个或多个实现方式示出并描述了本公开，但是本领域技术人员基于对本说明书和附图的阅读和理解将会想到等价变型和修改。本公开包括所有这样的修改和变型，并且仅由所附权利要求的范围限制。特别地关于由上述组件(例如元件、资源等)执行的各种功能，用于描述这样的组件的术语旨在对应于执行所述组件的指定功能(例如其在功能上是等价的)的任意组件(除非另外指示)，即使在结构上与执行本文所示的本公开的示范性实现方式中的功能的公开结构不等同。此外，尽管本公开的特定特征已经相对于若干实现方式中的仅一个被公开，但是这种特征可以与如可以对给定或特定应用而言是期望和有利的其他实现方式的一个或多个其他特征组合。而且，就术语“包括”、“具有”、“含有”或其变形被用在具体实施方式或权利要求中而言，这样的术语旨在以与术语“包含”相似的方式包括。
本发明实施例中的各功能单元可以集成在一个处理模块中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。所述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时，也可以存储在一个计算机可读取存储介质中。上述提到的存储介质可以是只读存储器，磁盘或光盘等。上述的各装置或***，可以执行相应方法实施例中的方法。
综上所述，虽然本发明已以优选实施例揭露如上，但上述优选实施例并非用以限制本发明，本领域的普通技术人员，在不脱离本发明的精神和范围内，均可作各种更动与润饰，因此本发明的保护范围以权利要求界定的范围为准。

Claims (14)

  1. 一种高动态范围视频拍摄方法，用于多镜头全景相机进行高动态范围视频的拍摄，所述多镜头全景相机具有多个不同拍摄方向的拍摄镜头，且相邻的拍摄镜头的拍摄图像之间具有重叠区域，其中所述高动态范围视频拍摄方法包括：
    使用所述多镜头全景相机的所述拍摄镜头获取对应的拍摄图像，其中相邻的拍摄镜头的拍摄图像的曝光强度不同；
    根据所述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重；
    基于每个所述拍摄图像的图像像素块的图像合成权重，将所有的拍摄图像进行合成，以形成高动态范围全景拍摄图像；
    对所述高动态范围全景拍摄图像进行亮度连续化处理，以得到处理后的高动态范围全景拍摄图像；以及
    将所有的所述处理后的高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频；
    所述根据所述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重的步骤包括：
    w＝a*L1+b*L2-c*L3；
    其中w为所述图像像素块的图像合成权重，L1为所述图像像素块的像素色彩灰阶变化量，L2为所述图像像素块的像素色彩灰阶差异量，L3为所述图像色彩灰阶偏移量，a为像素色彩灰阶变化量权重系数，b为像素色彩灰阶差异量权重系数，c为图像色彩灰阶偏移量权重系数；
    其中所述对所述高动态范围全景拍摄图像进行亮度连续化处理的步骤包括：
    Figure PCTCN2018095274-appb-100001
    Figure PCTCN2018095274-appb-100002
    其中N为拍摄镜头的数量，I表示拍摄图像中的图像像素块，T表示所述高动态范围全景拍摄图像中的图像像素块，O表示所述亮度连续化处理后的高动态范围全景拍摄图像中的图像像素块，
    Figure PCTCN2018095274-appb-100003
    为第n个拍摄镜头的第k个图像像素块的图像合成权重，v k(I n)为第n个拍摄镜头的第k个图像像素块的像素彩色灰阶值；v k(T)为所述高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值；v k(O)为所述亮度连续化处理后的高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值。
  2. 根据权利要求1所述的高动态范围视频拍摄方法，其中将曝光强度设定为高曝光强度以及低曝光强度，当所述拍摄镜头为高曝光强度时，相邻的拍摄镜头均为低曝光强度；当所述拍摄镜头为低曝光强度时， 相邻的拍摄镜头均为高曝光强度。
  3. 根据权利要求1所述的高动态范围视频拍摄方法，其中所述根据所述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重的步骤还包括：
    对所述图像合成权重进行归一化操作。
  4. 一种高动态范围视频拍摄方法，用于多镜头全景相机进行高动态范围视频的拍摄，所述多镜头全景相机具有多个不同拍摄方向的拍摄镜头，且相邻的拍摄镜头的拍摄图像之间具有重叠区域，其中所述高动态范围视频拍摄方法包括：
    使用所述多镜头全景相机的所述拍摄镜头获取对应的拍摄图像，其中相邻的拍摄镜头的拍摄图像的曝光强度不同；
    根据所述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重；
    基于每个所述拍摄图像的图像像素块的图像合成权重，将所有的拍摄图像进行合成，以形成高动态范围全景拍摄图像；以及
    将所有的所述高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频。
  5. 根据权利要求4所述的高动态范围视频拍摄方法，其中将曝光强度设定为高曝光强度以及低曝光强度，当所述拍摄镜头为高曝光强度时，相邻的拍摄镜头均为低曝光强度；当所述拍摄镜头为低曝光强度时，相邻的拍摄镜头均为高曝光强度。
  6. 根据权利要求4所述的高动态范围视频拍摄方法，其中所述根据所述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重的步骤包括：
    w＝a*L1+b*L2-c*L3；
    其中w为所述图像像素块的图像合成权重，L1为所述图像像素块的像素色彩灰阶变化量，L2为所述图像像素块的像素色彩灰阶差异量，L3为所述图像色彩灰阶偏移量，a为像素色彩灰阶变化量权重系数，b为像素色彩灰阶差异量权重系数，c为图像色彩灰阶偏移量权重系数。
  7. 根据权利要求4所述的高动态范围视频拍摄方法，其中所述形成所述高动态范围全景拍摄图像的步骤之后还包括：
    对所述高动态范围全景拍摄图像进行亮度连续化处理，以得到处理后的高动态范围全景拍摄图像；
    所述将所有的所述高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频的步骤为：
    将所有的所述处理后的高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频。
  8. 根据权利要求7所述的高动态范围视频拍摄方法，其中所述对所述高动态范围全景拍摄图像进行亮度连续化处理的步骤包括：
    Figure PCTCN2018095274-appb-100004
    Figure PCTCN2018095274-appb-100005
    其中N为拍摄镜头的数量，I表示拍摄图像中的图像像素块，T表示所述高动态范围全景拍摄图像中的图像像素块，O表示所述亮度连续化处理后的高动态范围全景拍摄图像中的图像像素块，
    Figure PCTCN2018095274-appb-100006
    为第n个拍摄镜头的第k个图像像素块的图像合成权重，v k(I n)为第n个拍摄镜头的第k个图像像素块的像素彩色灰阶值；v k(T)为所述高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值；v k(O)为所述亮度连续化处理后的高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值。
  9. 根据权利要求6所述的高动态范围视频拍摄方法，其中所述根据所述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重的步骤还包括：
    对所述图像合成权重进行归一化操作。
  10. 一种高动态范围视频拍摄装置，其设置在多镜头全景相机中，用于所述多镜头全景相机进行高动态范围视频的拍摄，所述多镜头全景相机具有多个不同拍摄方向的拍摄镜头，且相邻的拍摄镜头的拍摄图像之间具有重叠区域，其中所述高动态范围视频拍摄装置包括：
    拍摄图像获取模块，用于使用所述多镜头全景相机的所述拍摄镜头获取对应的拍摄图像，其中相邻的拍摄镜头的拍摄图像的曝光强度不同；
    图像合成权重确定模块，用于根据所述拍摄图像的图像像素块的像素色彩灰阶变化量、像素色彩灰阶差异量以及像素色彩灰阶偏移量确定对应的图像像素块的图像合成权重；
    图像合成模块，用于基于每个所述拍摄图像的图像像素块的图像合成权重，将所有的拍摄图像进行合成，以形成高动态范围全景拍摄图像；以及
    图像拼接模块，用于将所有的所述高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频。
  11. 根据权利要求10所述的高动态范围视频拍摄装置，其中将曝光强度设定为高曝光强度以及低曝光强度，当所述拍摄镜头为高曝光强度时，相邻的拍摄镜头均为低曝光强度；当所述拍摄镜头为低曝光强度时，相邻的拍摄镜头均为高曝光强度。
  12. 根据权利要求10所述的高动态范围视频拍摄装置，其中图像合成权重确定模块根据以下公式获取图像合成权重：
    w＝a*L1+b*L2-c*L3；
    其中w为所述图像像素块的图像合成权重，L1为所述图像像素块的像素色彩灰阶变化量，L2为所述图像像素块的像素色彩灰阶差异量，L3为所述图像色彩灰阶偏移量，a为像素色彩灰阶变化量权重系数，b 为像素色彩灰阶差异量权重系数，c为图像色彩灰阶偏移量权重系数。
  13. 根据权利要求10所述的高动态范围视频拍摄装置，其中所述高动态范围视频拍摄装置还包括：
    亮度连续化处理模块，用于对所述高动态范围全景拍摄图像进行亮度连续化处理，以得到处理后的高动态范围全景拍摄图像；
    所述图像拼接模块用于将所有的所述处理后的高动态范围全景拍摄图像进行拼接，以形成高动态范围全景视频。
  14. 根据权利要求13所述的高动态范围视频拍摄装置，其中所述亮度连续化处理模块根据以下公式对高动态范围全景拍摄图像进行亮度连续化处理：
    Figure PCTCN2018095274-appb-100007
    Figure PCTCN2018095274-appb-100008
    其中N为拍摄镜头的数量，I表示拍摄图像中的图像像素块，T表示所述高动态范围全景拍摄图像中的图像像素块，O表示所述亮度连续化处理后的高动态范围全景拍摄图像中的图像像素块，
    Figure PCTCN2018095274-appb-100009
    为第n个拍摄镜头的第k个图像像素块的图像合成权重，v k(I n)为第n个拍摄镜头的第k个图像像素块的像素彩色灰阶值；v k(T)为所述高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值；v k(O)为所述亮度连续化处理后的高动态范围全景拍摄图像的第k个图像像素块的像素彩色灰阶值。
PCT/CN2018/095274 2017-10-23 2018-07-11 高动态范围视频拍摄方法及拍摄装置 WO2019080543A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US16/758,405 US11070741B2 (en) 2017-10-23 2018-07-11 High dynamic range video shooting method and device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710994861.X 2017-10-23
CN201710994861.XA CN107800979B (zh) 2017-10-23 2017-10-23 高动态范围视频拍摄方法及拍摄装置

Publications (1)

Publication Number Publication Date
WO2019080543A1 true WO2019080543A1 (zh) 2019-05-02

Family

ID=61534255

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/095274 WO2019080543A1 (zh) 2017-10-23 2018-07-11 高动态范围视频拍摄方法及拍摄装置

Country Status (3)

Country Link
US (1) US11070741B2 (zh)
CN (1) CN107800979B (zh)
WO (1) WO2019080543A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112243091A (zh) * 2020-10-16 2021-01-19 微创（上海）医疗机器人有限公司 三维内窥镜***、控制方法和存储介质

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107800979B (zh) 2017-10-23 2019-06-28 深圳看到科技有限公司 高动态范围视频拍摄方法及拍摄装置
CN109005368B (zh) * 2018-10-15 2020-07-31 Oppo广东移动通信有限公司 一种高动态范围图像的生成方法、移动终端及存储介质
CN110166707B (zh) * 2019-06-13 2020-09-25 Oppo广东移动通信有限公司 图像处理方法、装置、电子设备以及存储介质
CN110581956A (zh) * 2019-08-26 2019-12-17 Oppo广东移动通信有限公司 图像处理方法、装置、存储介质及电子设备
CN110519483B (zh) * 2019-08-29 2023-08-22 圆周率科技(常州)有限公司 一种测试全景相机多镜头帧同步***
CN111311532B (zh) * 2020-03-26 2022-11-11 深圳市商汤科技有限公司 图像处理方法及装置、电子设备、存储介质
US11120606B1 (en) * 2020-04-24 2021-09-14 Electronic Arts Inc. Systems and methods for image texture uniformization for multiview object capture
CN111836017A (zh) * 2020-07-15 2020-10-27 广东欧谱曼迪科技有限公司 一种超长景深超宽动态内窥摄像***及方法
CN112399092A (zh) * 2020-10-27 2021-02-23 维沃移动通信有限公司 拍摄方法、装置和电子设备
CN116248895B (zh) * 2023-05-06 2023-07-21 上海扬谷网络科技有限公司 虚拟现实全景漫游的视频云转码方法及***

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1805551A (zh) * 2005-01-11 2006-07-19 吴水超 一种多摄像头实现宽动态范围图像采集的简易方法
US20100150473A1 (en) * 2008-12-16 2010-06-17 Jae-Hyun Kwon Apparatus and method for blending multiple images
CN102905058A (zh) * 2011-07-28 2013-01-30 三星电子株式会社 产生去除了重影模糊的高动态范围图像的设备和方法
CN106162024A (zh) * 2016-08-02 2016-11-23 乐视控股（北京）有限公司 照片处理方法及装置
CN107800979A (zh) * 2017-10-23 2018-03-13 深圳看到科技有限公司 高动态范围视频拍摄方法及拍摄装置

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040100565A1 (en) * 2002-11-22 2004-05-27 Eastman Kodak Company Method and system for generating images used in extended range panorama composition
CN101963751B (zh) * 2010-08-19 2011-11-30 西北工业大学 高分辨率实时全景高动态范围图像获取装置及方法
CN102420944B (zh) * 2011-04-25 2013-10-16 展讯通信（上海）有限公司 一种高动态范围图像合成方法及装置
US9955084B1 (en) * 2013-05-23 2018-04-24 Oliver Markus Haynold HDR video camera
JP6555863B2 (ja) * 2013-12-25 2019-08-07 キヤノン株式会社 撮像装置及び撮像装置の制御方法
US9485435B2 (en) * 2014-04-22 2016-11-01 Shih-Chieh Huang Device for synthesizing high dynamic range image based on per-pixel exposure mapping and method thereof
CN105469375B (zh) * 2014-08-28 2021-09-07 北京三星通信技术研究有限公司 处理高动态范围全景图的方法和装置
US9613408B2 (en) * 2014-09-25 2017-04-04 Intel Corporation High dynamic range image composition using multiple images
US9282256B1 (en) * 2014-12-22 2016-03-08 Omnivision Technologies, Inc. System and method for HDR imaging
US10186023B2 (en) * 2016-01-25 2019-01-22 Qualcomm Incorporated Unified multi-image fusion approach
JP6852411B2 (ja) * 2017-01-19 2021-03-31 ソニー株式会社 映像信号処理装置、映像信号処理方法およびプログラム
US10638052B2 (en) * 2017-04-12 2020-04-28 Samsung Electronics Co., Ltd. Method and apparatus for generating HDR images

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1805551A (zh) * 2005-01-11 2006-07-19 吴水超 一种多摄像头实现宽动态范围图像采集的简易方法
US20100150473A1 (en) * 2008-12-16 2010-06-17 Jae-Hyun Kwon Apparatus and method for blending multiple images
CN102905058A (zh) * 2011-07-28 2013-01-30 三星电子株式会社 产生去除了重影模糊的高动态范围图像的设备和方法
CN106162024A (zh) * 2016-08-02 2016-11-23 乐视控股（北京）有限公司 照片处理方法及装置
CN107800979A (zh) * 2017-10-23 2018-03-13 深圳看到科技有限公司 高动态范围视频拍摄方法及拍摄装置

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112243091A (zh) * 2020-10-16 2021-01-19 微创（上海）医疗机器人有限公司 三维内窥镜***、控制方法和存储介质
CN112243091B (zh) * 2020-10-16 2022-12-16 上海微创医疗机器人（集团）股份有限公司 三维内窥镜***、控制方法和存储介质

Also Published As

Publication number Publication date
CN107800979A (zh) 2018-03-13
US20200344399A1 (en) 2020-10-29
CN107800979B (zh) 2019-06-28
US11070741B2 (en) 2021-07-20

Similar Documents

Publication Publication Date Title
WO2019080543A1 (zh) 高动态范围视频拍摄方法及拍摄装置
TWI527448B (zh) 成像設備，影像處理方法，及用以記錄程式於其上之記錄媒體
WO2020207239A1 (en) Method and apparatus for image processing
WO2020034737A1 (zh) 成像控制方法、装置、电子设备以及计算机可读存储介质
EP3896955A1 (en) Image acquisition method, electronic device, and computer-readable storage medium
US20180288311A1 (en) Combining images when a face is present
CN109194877A (zh) 图像补偿方法和装置、计算机可读存储介质和电子设备
US10013747B2 (en) Image processing method, image processing apparatus and display apparatus
CN106063249A (zh) 摄像装置及其控制方法以及计算机可读记录介质
JP6720881B2 (ja) 画像処理装置及び画像処理方法
WO2020034701A1 (zh) 成像控制方法、装置、电子设备以及可读存储介质
TWI383673B (zh) 測光權重調整方法及其裝置
US20180139369A1 (en) Backlit face detection
CN110213494A (zh) 拍摄方法和装置、电子设备、计算机可读存储介质
JP2021505086A (ja) 画像処理方法および画像処理装置、コンピュータ可読記憶媒体ならびにコンピュータ機器
CN110290325A (zh) 图像处理方法、装置、存储介质及电子设备
JP2013162516A (ja) 画像自動補正のための構図に基づく露出測定方法及び装置
US10863103B2 (en) Setting apparatus, setting method, and storage medium
JP2010231757A (ja) Ｈｄｒ画像生成方法および同方法を用いるデジタル撮像素子
WO2021218536A1 (zh) 一种高动态范围图像合成方法和电子设备
JP2014179920A (ja) 撮像装置及びその制御方法、プログラム、並びに記憶媒体
US9900503B1 (en) Methods to automatically fix flash reflection at capture time
JP6157274B2 (ja) 撮像装置、情報処理方法及びプログラム
Li et al. Improving high dynamic range image based light measurement
JP2014155000A (ja) 画像処理装置、その制御方法、および制御プログラム

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: 18871541

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18871541

Country of ref document: EP

Kind code of ref document: A1