Breathing flicker removal processing method and device for movie picture and storage medium

Abstract

The embodiment of the invention provides a breathing flicker removal processing method and device for a movie picture and a storage medium. The method comprises the following steps: acquiring a to-be-processed film clip subjected to flicker removal processing; wherein the to-be-processed film clip after the flicker removal processing is obtained by performing deviation correction processing on the exposure abnormal frame; the exposure abnormal frame is determined from a plurality of film frames of the to-be-processed film clip according to the local space information capacity; the local space information capacity is used for representing regional energy distribution of the film frame; according to the local space information capacity of the to-be-processed film clip after the flicker removal processing, performing similarity division processing on the to-be-processed film clip after the flicker removal processing to obtain at least one to-be-processed sub-clip; and determining a standard frame of the to-be-processed sub-fragment, and performing optimization processing on the non-standard frame of the to-be-processed sub-fragment according to the standard frame to obtain the to-be-processed film fragment after breathing flicker removal processing. The method is used for achieving the effect of improving the film watching experience of the user.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular, to a method, device, and storage medium for removing breathing flicker from film images. Background Art

[0002] After correcting the exposure abnormal frames in the film segment to be processed to obtain the film segment to be processed after de-flicker processing, there will be a breathing flicker phenomenon in the processed film segment. Among them, the breathing flicker phenomenon refers to the phenomenon that in multiple consecutive film frames in the processed film segment, there will be a gradual change in brightness or color, which will affect the user's film viewing experience.

[0003] Therefore, there is an urgent need for a method for removing breathing flicker to improve the film viewing experience. Summary of the Invention

[0004] Embodiments of this application provide a method, device, and storage medium for removing breathing flicker from film images, so as to achieve the effect of improving the user's film viewing experience.

[0005] In a first aspect, embodiments of this application provide a method for removing breathing flicker from film images, including:

[0006] Obtain the film segment to be processed after de-flicker processing; wherein, the film segment to be processed is a segment of film images in the same film scene; the film segment to be processed after de-flicker processing is obtained by correcting the exposure abnormal frames; the exposure abnormal frames are determined from multiple film frames of the film segment to be processed according to the local spatial information capacity; the local spatial information capacity is used to characterize the regional energy distribution of the film frame;

[0007] According to the local spatial information capacity of the film segment to be processed after de-flicker processing, perform similarity division processing on the film segment to be processed after de-flicker processing to obtain at least one sub-segment to be processed;

[0008] Determine the standard frame of the sub-segment to be processed, and optimize the non-standard frames of the sub-segment to be processed according to the standard frame to obtain the film segment to be processed after breathing flicker removal processing.

[0009] In a possible implementation, the local space information capacity includes at least one local space information sub-capacity; wherein, the local space information sub-capacity is used to characterize the regional energy distribution of pixel blocks in the video frame; according to the local space information capacity of the processed video clip after de-flickering, perform a similarity division process on the processed video clip after de-flickering to obtain at least one sub-video clip to be processed, including: determining the similarity between every two video frames in the processed video clip after de-flickering according to at least one of the local space information sub-capacities included in the local space information capacity; performing a similarity division process on the processed video clip after de-flickering according to the similarity between every two video frames to obtain at least one sub-video clip to be processed.

[0010] In a possible implementation, determining the similarity between every two video frames in the processed video clip after de-flickering according to at least one of the local space information sub-capacities included in the local space information capacity includes: performing feature extraction processing on at least one of the local space information sub-capacities included in the local space information capacity to obtain the local space information sub-capacity feature vectors of each video frame;

[0011] Performing cosine similarity calculation processing on the local space information sub-capacity feature vectors of each video frame to obtain the similarity between every two video frames.

[0012] In a possible implementation, determining the similarity between every two video frames in the processed video clip after de-flickering according to at least one of the local space information sub-capacities included in the local space information capacity includes: performing feature extraction processing on at least one of the local space information sub-capacities included in the local space information capacity to obtain the local space information sub-capacity feature vectors of each video frame; obtaining the information entropy feature vectors of each video frame, and performing splicing processing on the local space information sub-capacity feature vectors and the information entropy feature vectors to obtain the integrated information feature vectors of each video frame; performing cosine similarity calculation processing on the integrated information feature vectors of each video frame to obtain the similarity between every two video frames.

[0013] In a possible implementation, performing a similarity division process on the processed video clip after de-flickering according to the similarity between every two video frames to obtain at least one sub-video clip to be processed includes: determining at least one video segmentation point based on dynamic programming technology according to the similarity between every two video frames; wherein, the video segmentation point is used to divide two sub-video clips to be processed; performing segmentation processing on the video clip to be processed according to the at least one video segmentation point to obtain the at least one sub-video clip to be processed.

[0014] In a possible implementation, determining the standard frame of the sub - fragment to be processed includes: obtaining at least one normalized histogram of each video frame in the sub - fragment to be processed; calculating the global similarity value of each video frame in the sub - fragment to be processed according to the at least one normalized histogram; wherein, the global similarity value is the sum of the similarity values between a video frame and other video frames; if there is only one video frame corresponding to the maximum value among the global similarity values of each video frame, determining this video frame as the standard frame; if there are multiple video frames corresponding to the maximum value among the global similarity values of each video frame, determining the video frame with the middle position in the time series among the multiple video frames as the standard frame.

[0015] In a possible implementation, optimizing the non - standard frames of the sub - fragment to be processed according to the standard frame to obtain the sub - fragment of the video to be processed after breathing and flickering removal includes: obtaining at least one normalized histogram of each video frame in the sub - fragment to be processed; calculating the normalized histogram cumulative distribution function value of each video frame in the sub - fragment to be processed according to the at least one normalized histogram; performing a mapping process on the normalized histogram cumulative distribution function value of each non - standard frame according to the normalized histogram cumulative distribution function value of the standard frame to obtain the sub - fragment of the video to be processed after breathing and flickering removal.

[0016] In a possible implementation, after optimizing the non - standard frames of the sub - fragment to be processed according to the standard frame, the method further includes: calculating the brightness histogram cumulative distribution function value of the standard frame of each optimized sub - fragment to be processed, and determining the brightness mapping relationship between adjacent optimized sub - fragments to be processed based on the brightness histogram cumulative distribution function value; setting an overlapping area between adjacent optimized sub - fragments to be processed, and performing brightness interpolation adjustment on each video frame in the overlapping area based on the brightness mapping relationship to obtain two sets of overlapping areas after brightness interpolation adjustment; wherein, the overlapping area includes all boundary frames of adjacent optimized sub - fragments to be processed; generating the value of the transparency channel corresponding to the overlapping area, and performing weighted fusion processing on the pixel brightness values of each video frame in the two sets of overlapping areas after brightness interpolation adjustment according to the value of the transparency channel to obtain the sub - fragment of the video to be processed after breathing and flickering removal.

[0017] In a second aspect, an embodiment of the present application provides a breathing and flickering removal processing device for a video picture, including:

[0018] An acquisition module, configured to acquire a to-be-processed video clip after de-flickering processing; wherein, the to-be-processed video clip is a clip of video frames under the same video scene; the to-be-processed video clip after de-flickering processing is obtained by rectifying an exposure-abnormal frame; the exposure-abnormal frame is determined from multiple video frames of the to-be-processed video clip according to the local space information capacity; the local space information capacity is used to characterize the regional energy distribution of the video frame.

[0019] A first processing module, configured to perform a similarity division process on the to-be-processed video clip after de-flickering processing according to the local space information capacity of the to-be-processed video clip after de-flickering processing, to obtain at least one to-be-processed sub-clip.

[0020] A second processing module, configured to determine a standard frame of the to-be-processed sub-clip, and optimize non-standard frames of the to-be-processed sub-clip according to the standard frame, to obtain a to-be-processed video clip after breathing flicker removal processing.

[0021] In a possible implementation manner, the local space information capacity includes at least one local space information sub-capacity; wherein, the local space information sub-capacity is used to characterize the regional energy distribution of a pixel block in the video frame; the first processing module is specifically configured to determine the similarity between every two video frames in the to-be-processed video clip after de-flickering processing according to at least one of the local space information sub-capacities included in the local space information capacity; and perform a similarity division process on the to-be-processed video clip after de-flickering processing according to the similarity between every two video frames, to obtain at least one to-be-processed sub-clip.

[0022] In a possible implementation manner, the first processing module is further specifically configured to perform feature extraction processing on at least one of the local space information sub-capacities included in the local space information capacity, to obtain a local space information sub-capacity feature vector of each video frame; and perform a cosine similarity calculation process on the local space information sub-capacity feature vectors of each video frame, to obtain the similarity between every two video frames.

[0023] In a possible implementation manner, the first processing module is further specifically configured to perform feature extraction processing on at least one of the local space information sub-capacities included in the local space information capacity, to obtain a local space information sub-capacity feature vector of each video frame; obtain an information entropy feature vector of each video frame, and splice the local space information sub-capacity feature vector and the information entropy feature vector, to obtain an integrated information feature vector of each video frame; and perform a cosine similarity calculation process on the integrated information feature vectors of each video frame, to obtain the similarity between every two video frames.

[0024] In a possible implementation, the first processing module is further specifically configured to determine at least one video segmentation point based on dynamic programming technology according to the similarity between every two video frames; wherein, the video segmentation point is used to divide two of the to-be-processed segments; and segment the to-be-processed video segment according to the at least one video segmentation point to obtain the at least one to-be-processed sub-segment.

[0025] In a possible implementation, the second processing module is specifically configured to obtain at least one normalized histogram of each video frame in the to-be-processed sub-segment; calculate the global similarity value of each video frame in the to-be-processed sub-segment according to the at least one normalized histogram; wherein, the global similarity value is the sum value of the similarity values between a video frame and other video frames; if there is only one video frame corresponding to the maximum value among the global similarity values of each video frame, determine this video frame as the standard frame; if there are multiple video frames corresponding to the maximum value among the global similarity values of each video frame, determine the video frame at the middle position in the time series among the multiple video frames as the standard frame.

[0026] In a possible implementation, the second processing module is further specifically configured to obtain at least one normalized histogram of each video frame in the to-be-processed sub-segment; calculate the normalized histogram cumulative distribution function value of each video frame in the to-be-processed sub-segment according to the at least one normalized histogram; perform a mapping process on the normalized histogram cumulative distribution function value of each non-standard frame according to the normalized histogram cumulative distribution function value of the standard frame to obtain the to-be-processed video segment after breathing and flickering removal processing.

[0027] In a possible implementation, the second processing module is further configured to calculate the luminance histogram cumulative distribution function value of the standard frame of each optimized to-be-processed sub-segment, and determine the luminance mapping relationship between adjacent optimized to-be-processed sub-segments based on the luminance histogram cumulative distribution function value; set an overlapping area between adjacent optimized to-be-processed sub-segments, and perform luminance interpolation adjustment on each video frame in the overlapping area based on the luminance mapping relationship to obtain two sets of overlapping areas after luminance interpolation adjustment; wherein, the overlapping area includes all boundary frames of adjacent optimized to-be-processed sub-segments; generate the value of the transparency channel corresponding to the overlapping area, and perform weighted fusion processing on the pixel luminance values of each video frame in the two sets of overlapping areas after luminance interpolation adjustment according to the value of the transparency channel to obtain the to-be-processed video segment after breathing and flickering removal processing.

[0028] In a third aspect, an embodiment of the present application provides an electronic device, including: a memory, a processor;

[0029] The memory stores computer-executable instructions;

[0030] The processor executes the computer-executable instructions stored in the memory, such that the processor performs the above first aspect and / or various possible implementation manners of the first aspect.

[0031] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the above first aspect and / or various possible implementation manners of the first aspect.

[0032] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the above first aspect and / or various possible implementation manners of the first aspect.

[0033] The method, device, and storage medium for removing breathing flicker from a film picture provided by the embodiments of the present application can efficiently and accurately remove the breathing flicker phenomenon of the film by dividing sub-fragments to be processed and optimizing non-standard frames through standard frames of the sub-fragments to be processed, improving the user's film viewing experience. Among them, through the local spatial information capacity, the sub-fragments to be processed can be accurately divided from the processed film fragments after de-flicker processing, further improving the accuracy of film processing, thereby improving the user's film viewing experience. Based on the above description, the film processing method provided by the present application can improve the user's film viewing experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0035] Figure 1 Schematic flowchart of the method for removing breathing flicker from a film picture provided by the present application Figure 1 ;

[0036] Figure 2 Schematic flowchart of the method for removing breathing flicker from a film picture provided by the present application Figure 2 ;

[0037] Figure 3 Schematic flowchart of the method for removing breathing flicker from a film picture provided by the present application Figure 3 ;

[0038] Figure 4 Schematic flowchart of the method for removing breathing flicker from a film picture provided by the present application Figure 4 ;

[0039] Figure 5Schematic flowchart of the method for removing breathing flicker from the video frame provided by this application Figure 5 ;

[0040] Figure 6 Schematic structural diagram of the device for removing breathing flicker from the video frame provided by this application;

[0041] Figure 7 Schematic structural diagram of the electronic device provided by this application.

[0042] Through the above-mentioned drawings, specific embodiments of this application have been shown, and there will be more detailed descriptions hereinafter. These drawings and text descriptions are not intended to limit the scope of the concept of this application in any way, but to illustrate the concept of this application to those skilled in the art by referring to specific embodiments. Detailed Description of Specific Embodiments

[0043] Exemplary embodiments will be described in detail herein, and examples thereof are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. On the contrary, they are merely examples of devices and methods consistent with some aspects of this application as detailed in the appended claims.

[0044] In the prior art, after rectifying the exposure abnormal frames in the video clip to be processed to obtain the video clip to be processed after de-flicker processing, there will be a breathing flicker phenomenon in the processed video clip. Among them, the breathing flicker phenomenon refers to the phenomenon that in multiple consecutive video frames in the processed video clip, there will be a gradual change in brightness or color, and this phenomenon will also affect the user's video viewing experience. Therefore, there is an urgent need for a method for removing breathing flicker to improve the video viewing experience.

[0045] The method for removing breathing flicker from the video frame provided by this application can efficiently and accurately remove the breathing flicker phenomenon of the video and improve the user's video viewing experience by dividing the video clip to be processed and optimizing the non-standard frames through the standard frames of the video clip to be processed. Among them, through the local spatial information capacity, the video clip to be processed can be accurately divided from the video clip to be processed after de-flicker processing, further improving the accuracy of video processing, thereby improving the user's video viewing experience. Based on the above description, the video processing method provided by this application can improve the user's video viewing experience.

[0046] The technical solution of the present application and how the technical solution of the present application solves the above technical problems will be described in detail below with specific embodiments. These several specific embodiments below can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0047] Figure 1 Schematic flow of the method for removing breathing flicker from the video frame provided by the present application Figure 1 , as Figure 1 shown, the method includes:

[0048] Step S101, obtain a to-be-processed video clip after de-flicker processing.

[0049] Specifically, a to-be-processed video clip after de-flicker processing can be obtained. Among them, the to-be-processed video clip is a clip of video frames in the same video scene. The to-be-processed video clip after de-flicker processing is obtained by correcting the exposure abnormal frames. The exposure abnormal frames are determined from multiple video frames of the to-be-processed video clip according to the local space information capacity. The local space information capacity is used to characterize the regional energy distribution of the video frames.

[0050] Specifically, the present application does not limit the process of obtaining the to-be-processed video clip after de-flicker processing. Optionally, a to-be-processed video clip can be obtained; among them, the to-be-processed video clip is a clip of video frames in the same video scene; the to-be-processed video clip includes multiple video frames; determine the local space information capacity of the video frames, and determine the exposure abnormal frames from the multiple video frames according to the local space information capacity; among them, the local space information capacity is used to characterize the regional energy distribution of the video frames; correct the exposure abnormal frames according to the pixel values of the adjacent frames of the exposure abnormal frames to obtain the to-be-processed video after de-flicker processing.

[0051] Specifically, during the process of de-flicker processing of the video, if the to-be-processed video is a clip of video frames in different video scenes, there will be a problem that the determination of abnormal frames is inaccurate due to different video scenes to which each video frame belongs, which further leads to the inaccuracy of the de-flicker processing process of the video. Therefore, a to-be-processed video clip can be obtained. Among them, the to-be-processed video clip is a clip of video frames in the same video scene. The to-be-processed video clip includes multiple video frames.

[0052] Specifically, the present application does not limit the process of obtaining the film segment to be processed. Optionally, a complete film can be obtained; where the complete film is a film segment including at least one film segment to be processed; based on the deep feature extraction technology and the shallow feature extraction technology, the similarity between each adjacent film frame in the complete film is determined; according to the similarity between each adjacent film frame in the complete film, the complete film is divided to obtain at least one film segment to be processed.

[0053] Specifically, after obtaining the film segment to be processed, the local spatial information capacity of each film frame in the film segment to be processed can be determined.

[0054] Among them, in an optical imaging system, objects outside the depth of field range will produce defocus blur. When the object is far from the focal plane, its spatial high-frequency components, that is, detail information, will decay non-linearly with the increase of the defocus distance, showing a change in the band-limited characteristic of the spatial frequency. A more accurate description should be "Local Spatial Information Capacity" (LSIC for short). Specifically, for the near-view region, a unit pixel corresponds to a smaller physical scale and has a higher spatial sampling density, that is, a high local spatial information capacity; for the far-view region, a unit pixel covers a larger physical scale, resulting in spatial aliasing, that is, a low local spatial information capacity.

[0055] Among them, the local spatial information capacity is used to characterize the regional energy distribution of the film frame. Specifically, based on the above description of the local spatial information capacity, the stronger the regional energy distribution of the film frame characterized by the local spatial information capacity, the higher the local spatial information capacity; conversely, the weaker the regional energy distribution of the film frame characterized by the local spatial information capacity, the lower the local spatial information capacity.

[0056] Specifically, the present application does not limit the process of determining the local spatial information capacity of the film frame. Optionally, the noise variance of the film frame can be obtained, and the spectral energy of each pixel block in the film frame can be obtained; according to the noise variance and the spectral energy of each pixel block, the local spatial information capacity is determined.

[0057] Specifically, after determining the local spatial information capacity of each film frame in the film segment to be processed, the exposure abnormal frame can be determined from multiple film frames according to the local spatial information capacity.

[0058] Among them, the exposure abnormal frame is the film frame to be corrected among multiple film frames of the film segment to be processed during the de-flickering process. Among them, the present application does not limit the number of determined exposure abnormal frames.

[0059] Specifically, this application does not limit the process of determining the exposure abnormal frame from multiple video frames according to the local space information capacity. Optionally, at least one local space information sub-capacity is included in the local space information capacity; wherein, the local space information sub-capacity is used to characterize the regional energy distribution of pixel blocks in the video frame; according to at least one local space information sub-capacity included in the local space information capacity, calculate the probability density of the local space information sub-capacity of the video frame; wherein, the probability density of the local space information sub-capacity indicates the proportion of the number of pixel blocks corresponding to each local space information sub-capacity to the total number of pixel blocks in the video frame; according to at least one local space information sub-capacity included in the local space information capacity, construct a heat map histogram of the video frame; determine whether the video frame is an exposure abnormal frame according to the proportion of the number of pixel blocks corresponding to each local space information sub-capacity in the video frame to the total number of pixel blocks in the video frame, and / or the heat map histogram of the video frame.

[0060] Specifically, after determining the exposure abnormal frame, the exposure abnormal frame can be corrected according to the pixel values of the adjacent frames of the exposure abnormal frame to obtain the video to be processed after de-flickering processing.

[0061] Specifically, this application does not limit the process of correcting the exposure abnormal frame according to the pixel values of the adjacent frames of the exposure abnormal frame to obtain the video to be processed after de-flickering processing. Optionally, regression calculation processing can be performed on the pixel values of the adjacent frames of the exposure abnormal frame to obtain the target pixel value of the exposure abnormal frame; correct the pixel values of the exposure abnormal frame according to the target pixel value of the exposure abnormal frame to obtain the exposure abnormal frame after correction processing; if the structural similarity index and peak signal-to-noise ratio between the exposure abnormal frame after correction processing and the exposure abnormal frame both meet the preset conditions, then replace the exposure abnormal frame with the exposure abnormal frame after correction processing; otherwise, increase the number of adjacent frames of the exposure abnormal frame and repeat the process of regression calculation processing and correction processing.

[0062] Among them, through the local space information capacity, the exposure abnormal frame can be accurately determined from the video segment to be processed, further improving the accuracy of the video to be processed after de-flickering processing, thereby improving the user's video viewing experience.

[0063] Among them, after performing correction processing on the exposure abnormal frame in the video segment to be processed based on the above-described process to obtain the video segment to be processed after de-flickering processing, there will be a breathing flicker phenomenon in the processed video segment to be processed. Among them, the breathing flicker phenomenon refers to the phenomenon that there will be a gradual change in brightness or color in multiple consecutive video frames in the processed video segment, and this phenomenon will also affect the user's video viewing experience.

[0064] Among them, the video segment to be processed after de-flickering processing can be expressed as: , where T represents the total number of video frames included in the to-be-processed video segment after de-flickering processing, and F t represents the t-th video frame, where (t = 1, 2, ⋯, T).

[0065] Step S102: Perform similarity division processing on the to-be-processed video segment after de-flickering processing according to the local spatial information capacity of the to-be-processed video segment after de-flickering processing, to obtain at least one to-be-processed sub-segment.

[0066] Specifically, in the segment of the video picture in the same video scene, if there is a video segment with a relatively high similarity, the change in brightness or color within this video segment should be relatively small. Therefore, during the process of removing breathing flicker, the video segments in the same video scene can be first divided into at least one video segment with a relatively high similarity. However, during the process of dividing the video segments in the same video scene, if there are abnormal exposure frames, there will be a problem that the division processing of the video segments is inaccurate, resulting in inaccurate removal of breathing flicker. Therefore, during the process of removing breathing flicker from the video segments in the same video scene, the abnormal exposure frames therein can be first corrected, that is, the to-be-processed video segment after de-flickering processing described in step S101 is generated.

[0067] Specifically, after obtaining the to-be-processed video segment after de-flickering processing, the to-be-processed video segment after de-flickering processing can be subjected to similarity division processing according to the local spatial information capacity of the to-be-processed video segment after de-flickering processing, to obtain at least one to-be-processed sub-segment. Among them, the to-be-processed sub-segment is a video segment with a relatively high similarity in the to-be-processed video segment after de-flickering processing.

[0068] Specifically, this application does not limit the process of performing similarity division processing on the to-be-processed video segment after de-flickering processing according to the local spatial information capacity of the to-be-processed video segment after de-flickering processing to obtain at least one to-be-processed sub-segment. Optionally, the local spatial information capacity includes at least one local spatial information sub-capacity; where the local spatial information sub-capacity is used to characterize the regional energy distribution of pixel blocks in the video frame; the similarity between every two video frames in the to-be-processed video segment after de-flickering processing can be determined according to at least one local spatial information sub-capacity included in the local spatial information capacity; and similarity division processing is performed on the to-be-processed video segment after de-flickering processing according to the similarity between every two video frames, to obtain at least one to-be-processed sub-segment.

[0069] Among them, the to-be-processed sub-segment can be expressed as: . Among them, the to-be-processed video segment after de-flickering processing includes K + 1 to-be-processed sub-segments. Among them, A set of video segmentation points of the video clip to be processed after de - flickering processing, where 。

[0070] Step S103: Determine the standard frame of the sub - clip to be processed, and optimize the non - standard frames of the sub - clip to be processed according to the standard frame, so as to obtain the video clip to be processed after respiration flickering removal processing.

[0071] Specifically, after obtaining a sub - clip to be processed, the standard frame of each sub - clip to be processed can be determined, and the non - standard frames of the corresponding sub - clip to be processed can be optimized according to the standard frame, so as to obtain the video clip to be processed after respiration flickering removal processing.

[0072] Among them, the standard frame is the video frame determined as the optimization standard in the optimization process among multiple video frames of the sub - clip to be processed. Specifically, the present application does not limit the process of determining the standard frame of the sub - clip to be processed. Optionally, at least one normalized histogram of each video frame in the sub - clip to be processed can be obtained; according to the at least one normalized histogram, calculate the global similarity value of each video frame in the sub - clip to be processed; among them, the global similarity value is the sum value of the similarity values between the video frame and other video frames; if the number of video frames corresponding to the maximum value in the global similarity values of each video frame is one, then determine this video frame as the standard frame; if the number of video frames corresponding to the maximum value in the global similarity values of each video frame is multiple, then determine the video frame located in the middle position in the time series among the multiple video frames as the standard frame. Among them, each sub - clip to be processed includes one standard frame.

[0073] Among them, the non - standard frame is the video frame to be optimized among multiple video frames of the sub - clip to be processed. Among them, other video frames in the sub - clip to be processed except the standard frame are all non - standard frames of this sub - clip to be processed.

[0074] Specifically, the present application does not limit the process of optimizing the non - standard frames of the corresponding sub - clip to be processed according to the standard frame to obtain the video clip to be processed after respiration flickering removal processing. Optionally, at least one normalized histogram of each video frame in the sub - clip to be processed can be obtained; according to the at least one normalized histogram, calculate the normalized histogram cumulative distribution function value of each video frame in the sub - clip to be processed; according to the normalized histogram cumulative distribution function value of the standard frame, perform mapping processing on the normalized histogram cumulative distribution function value of each non - standard frame to obtain the video clip to be processed after respiration flickering removal processing.

[0075] The method for removing breathing flicker from a video frame provided by an embodiment of the present application can efficiently and accurately remove the breathing flicker phenomenon of a video by dividing a sub-segment to be processed and optimizing non-standard frames through standard frames of the sub-segment to be processed, thereby improving the user's video viewing experience. Among them, through the local spatial information capacity, the sub-segments to be processed can be accurately divided from the processed video segment after de-flickering processing, further improving the accuracy of video processing, and thus improving the user's video viewing experience. Based on the above description, the video processing method provided by the present application can improve the user's video viewing experience.

[0076] Figure 2 is a flowchart of the method for removing breathing flicker from a video frame provided by the present application Figure 2 , such as Figure 2 shown. Based on the Figure 1 embodiment, the process of dividing the processed video segment after de-flickering processing into similarity segments according to the local spatial information capacity of the processed video segment after de-flickering processing to obtain at least one sub-segment to be processed is described in detail. The method includes:

[0077] Step S201: Determine the similarity between every two video frames in the processed video segment after de-flickering processing according to at least one local spatial information sub-capacity included in the local spatial information capacity.

[0078] Among them, the local spatial information capacity includes at least one local spatial information sub-capacity. The local spatial information sub-capacity is used to characterize the regional energy distribution of pixel blocks in a video frame.

[0079] Specifically, the number of local spatial information sub-capacities included in the local spatial information capacity corresponds to the number of pixel blocks included in the video frame. Specifically, the pixel block is a pixel block centered on the pixel point (i, j) in the video frame and with a preset size. In the present application, the preset size of the pixel blocks included in the video frame is not limited. Optionally, the preset size can be 8 pixels × 8 pixels.

[0080] Among them, the local spatial information capacity of a video frame can be expressed as: , where LSIC T represents the local spatial information sub-capacity, and T represents the number of local spatial information sub-capacities included in the local spatial information capacity.

[0081] Specifically, this application does not limit the process of determining the similarity between every two video frames in the to-be-processed video clip after de-flickering processing according to at least one local space information sub-capacity included in the local space information capacity. Optionally, determining the similarity between every two video frames in the to-be-processed video clip after de-flickering processing according to at least one local space information sub-capacity included in the local space information capacity includes:

[0082] Performing feature extraction processing on at least one local space information sub-capacity included in the local space information capacity to obtain the local space information sub-capacity feature vectors of each video frame.

[0083] Performing cosine similarity calculation processing on the local space information sub-capacity feature vectors of each video frame to obtain the similarity between every two video frames.

[0084] Specifically, in the process of performing feature extraction processing on at least one local space information sub-capacity included in the local space information capacity to obtain the local space information sub-capacity feature vectors of each video frame:

[0085] The value range [S min , S max of the local space information sub-capacity (i.e., the local stability index LSIC representing the regional energy distribution) can be obtained first, and it is divided into N segments, and each segment interval is (where ; secondly, for each video frame image, count the number of pixel blocks whose local space information sub-capacity values (LSIC values) of each pixel block fall within each interval, and perform normalization processing to obtain the LSIC feature component , where is the number of pixel blocks whose LSIC value in the t-th frame falls within the k-th interval, and is the total number of pixel blocks.

[0086] Specifically, the formula in the process of performing cosine similarity calculation processing on the local space information sub-capacity feature vectors of each video frame to obtain the similarity between every two video frames is as follows:

[0087]

[0088] Where is the dot product of vectors and , and are the norms of vectors and respectively, and the calculation formula is .

[0089] Among them, in the process of determining the similarity between every two video frames in the to-be-processed video clip after de-flickering processing according to at least one local space information sub-capacity included in the local space information capacity, through the process of feature extraction processing and cosine similarity calculation on at least one local space information sub-capacity included in the local space information capacity, the similarity between every two video frames in the to-be-processed video clip after de-flickering processing can be accurately and efficiently determined, further improving the efficiency and accuracy of video processing, thereby enhancing the user's video viewing experience.

[0090] Optionally, determining the similarity between every two video frames in the to-be-processed video clip after de-flickering processing according to at least one local space information sub-capacity included in the local space information capacity includes:

[0091] Performing feature extraction processing on at least one local space information sub-capacity included in the local space information capacity to obtain local space information sub-capacity feature vectors of each video frame;

[0092] Obtaining information entropy feature vectors of each video frame, and performing splicing processing on the local space information sub-capacity feature vectors and the information entropy feature vectors to obtain integrated information feature vectors of each video frame;

[0093] Performing cosine similarity calculation processing on the integrated information feature vectors of each video frame to obtain the similarity between every two video frames.

[0094] Specifically, in the process of calculating the information entropy feature vectors of each video frame, the global information entropy H of each video frame can be calculated first t (measuring the uncertainty of pixel gray scale distribution), and then normalizing it to interval to obtain the information entropy feature vector, that is, the information entropy feature component, and its expression is: , where H min and H max are the minimum value and the maximum value of the global information entropy of each video frame in the video clip, that is, the to-be-processed video clip after de-flickering processing, represents a constant to avoid the denominator being zero, for example, 10 6 .

[0095] Specifically, in the process of performing splicing processing on the information entropy feature vector and the local space information sub-capacity feature vector to obtain the integrated information feature of each video frame, the local space information sub-capacity feature vector and the information entropy feature vector can be spliced to form a local space information sub-capacity feature vector including regional energy distribution features and gray scale distribution complexity features .

[0096] Specifically, for the process of calculating the cosine similarity of the integrated information feature vectors of each video frame to obtain the similarity between every two video frames, reference can be made to the process of calculating the cosine similarity of the local space information sub-capacity feature vectors of each video frame described above to obtain the similarity between every two video frames, which will not be elaborated here.

[0097] Specifically, information entropy is used to measure the uncertainty of the pixel gray-scale distribution of a video frame. After calculation, it is used as an independent feature component and concatenated with the local stability index (LSIC) feature component that characterizes the regional energy distribution. In this way, the local space information sub-capacity feature vector not only reflects the regional energy distribution of pixel blocks in the video frame through the LSIC feature component (such as by counting the number of pixel points in different intervals of the LSIC value and normalizing to form a basic feature reflecting the characteristics of regional energy distribution), but also reflects the complexity of the pixel gray-scale distribution through the information entropy feature component (after normalizing the global information entropy and integrating it into the feature vector), forming a complete feature vector containing the complexity of gray-scale distribution and the characteristics of regional energy distribution, laying a comprehensive feature foundation for accurately measuring the similarity between video frames.

[0098] Among them, in the process of determining the similarity between every two video frames in the to-be-processed video segment after de-flickering based on at least one local space information sub-capacity included in the local space information capacity, by fusing the information entropy feature and the local space information sub-capacity feature, the similarity calculation simultaneously reflects the local stability of the video frame (the LSIC feature based on the regional energy distribution) and the global gray-scale distribution consistency (the gray-scale distribution uncertainty characterized by information entropy). This method more comprehensively and accurately depicts the similarity relationship between video frames: the stability of the regional energy distribution of pixel blocks within the frame is reflected based on the LSIC feature, and the complexity characteristics of the global gray-scale distribution of the frame are reflected based on the information entropy. The combination of the two enables the similarity calculation result to better represent the comprehensive similarity degree of the video frames in terms of brightness content and gray-scale distribution. This makes the subsequent video segment division more reasonable (such as when determining the segmentation point through dynamic programming, it can more accurately divide the sub-segments with basically the same brightness content and stable gray-scale distribution complexity), provides a more reliable basis for optimizing non-standard frames based on standard frames, and thus more efficiently and accurately removes the breathing flicker phenomenon of the video. This process improves the accuracy of video processing, enhances the naturalness of the transition between video frames, reduces the visual flicker caused by local brightness changes and gray-scale distribution fluctuations, and significantly improves the user viewing experience.

[0099] Step S202: Perform similarity division processing on the to-be-processed video segment after de-flickering according to the similarity between every two video frames to obtain at least one to-be-processed sub-segment.

[0100] Specifically, this application does not limit the process of performing similarity division processing on the to-be-processed video segment after de-flickering processing according to the similarity between every two video frames. Optionally, performing similarity division processing on the to-be-processed video segment after de-flickering processing according to the similarity between every two video frames to obtain at least one to-be-processed sub-segment includes:

[0101] Based on dynamic programming technology, determine at least one video segmentation point according to the similarity between every two video frames. Among them, the video segmentation point is used to divide two to-be-processed segments.

[0102] Perform segmentation processing on the to-be-processed video segment according to at least one video segmentation point to obtain at least one to-be-processed sub-segment.

[0103] Specifically, in the process of determining at least one video segmentation point based on dynamic programming technology according to the similarity between every two video frames, a target function can be set so that the similarity of each video frame in the to-be-processed sub-segment segmented by the video segmentation point is the highest when the target function is minimized. Among them, the formula of the target function can be:

[0104]

[0105] Among them, C(S) represents the value of the target function, |V k+1 | represents the number of video frames included in the to-be-processed sub-segment V k+1 , sim(i, j) represents the similarity between the i-th video frame and the j-th video frame. Among them, λ1 represents a trade-off coefficient used to balance the similarity within the to-be-processed sub-segment and the number of to-be-processed sub-segments.

[0106] Among them, the formula for determining at least one video segmentation point based on dynamic programming technology according to the similarity between every two video frames is as follows:

[0107]

[0108] Among them, dp[t] represents the minimum target function value of dividing the first t video frames into several to-be-processed sub-segments, λ2 represents a trade-off coefficient used to balance the similarity within the to-be-processed sub-segment and the number of to-be-processed sub-segments, and dp[1] = 0.

[0109] Specifically, in the process of determining the video segmentation point based on the above formula, processes such as an initialization process, an iterative calculation process, a final result determination process, and a backtracking path process can be involved.

[0110] Among them, in the process of performing similarity division processing on the to-be-processed video clip after de-flickering processing according to the similarity between every two video frames to obtain at least one to-be-processed sub-clip, the video segmentation point can be accurately determined based on the dynamic programming technique, which can improve the accuracy of the division of the to-be-processed sub-clip, thereby improving the accuracy of video processing and further enhancing the user's video viewing experience.

[0111] In the process provided by the embodiment of the present application of performing similarity division processing on the to-be-processed video clip after de-flickering processing according to the local space information capacity of the to-be-processed video clip after de-flickering processing to obtain at least one to-be-processed sub-clip, by determining the similarity between every two video frames in the to-be-processed video clip after de-flickering processing according to at least one local space information sub-capacity included in the local space information capacity, and performing similarity division processing on the to-be-processed video clip after de-flickering processing according to the similarity between every two video frames to obtain at least one to-be-processed sub-clip. Among them, by calculating the similarity between every two video frames in the to-be-processed video clip after de-flickering processing, the to-be-processed sub-clip can be accurately and efficiently divided, thereby improving the efficiency and accuracy of video processing and further enhancing the user's video viewing experience.

[0112] Figure 3 It is a flowchart illustration of the method for removing breathing flicker from the video frame provided by the present application. Figure 3 , such as Figure 3 shown. Based on the embodiment of Figure 1 or Figure 2 embodiment, the process of determining the standard frame of the to-be-processed sub-clip is described in detail. The method includes:

[0113] Step S301: Obtain at least one normalized histogram of each video frame in the to-be-processed sub-clip.

[0114] Specifically, at least one normalized histogram of each video frame in the to-be-processed sub-clip can be obtained.

[0115] Specifically, the present application does not limit the process of obtaining at least one normalized histogram of each video frame in the to-be-processed sub-clip. Optionally, each video frame in the to-be-processed sub-clip can be converted into the HSV color space, where the HSV color space includes the brightness color channel V, the hue color channel H, and the saturation color channel S; and then the normalized histograms of each color channel are calculated respectively.

[0116] Optionally, after obtaining at least one normalized histogram of each video frame in the to-be-processed sub-clip, Gaussian smoothing processing can be performed on the histogram to reduce the interference of noise on the similarity calculation. Optionally, the smoothing coefficient σ = 1.5.

[0117] Step S302: Calculate the global similarity value of each video frame in the sub - segment to be processed according to at least one normalized histogram.

[0118] Specifically, according to at least one normalized histogram obtained in step S301, the global similarity value of each video frame in the sub - segment to be processed can be calculated. Among them, the global similarity value is the sum of the similarity values between a video frame and other video frames.

[0119] Specifically, this application does not limit the process of calculating the global similarity value of each video frame in the sub - segment to be processed according to at least one normalized histogram. Optionally, the similarity matrix between every two video frames in the sub - segment to be processed can be calculated first, and then the global similarity value of each video frame in the sub - segment to be processed can be calculated according to the similarity matrix between every two video frames in the sub - segment to be processed.

[0120] Specifically, this application does not limit the process of calculating the similarity matrix between every two video frames in the sub - segment to be processed. Optionally, the similarity matrix between every two video frames in the sub - segment to be processed can be calculated according to the Bhattacharyya distance calculation method.

[0121] Among them, the calculation formula of the similarity matrix between every two video frames is as follows:

[0122]

[0123] Among them, d B (H i ,H j ) represents the similarity matrix between the i - th video frame and the j - th video frame, that is, the Bhattacharyya distance. Among them, the smaller the Bhattacharyya distance, the more similar the distribution of the histograms between the two video frames. Among them, H i and H j represent the normalized histograms of the i - th video frame and the j - th video frame. Among them, L represents the number of pixel values in the normalized histogram, for example, 256.

[0124] Specifically, the calculation formula of the global similarity value of each video frame is as follows:

[0125]

[0126] Among them, S(F i ) represents the global similarity value of the video frame F i , represents the smoothing constant, for example, 10 -3 , which is used to avoid the denominator being zero.

[0127] Among them, the above - described calculation of the similarity matrix and the global similarity value is for the similarity matrix and the global similarity value corresponding to the luminance - chrominance channels.

[0128] Among them, comprehensively consider the normalized histograms of each color channel described in step S301. For example, in the process of obtaining the similarity matrix between every two video frames, obtain the similarity matrix corresponding to the normalized histogram of each color channel. For example, in the process of calculating the global similarity value of each video frame, the similarity matrix corresponding to the normalized histogram of each obtained color channel can be comprehensively considered.

[0129] Step S303: If there is only one video frame corresponding to the maximum value among the global similarity values of each video frame, then determine this video frame as the standard frame.

[0130] Specifically, according to the calculation process in step S302, if it is determined that there is only one video frame corresponding to the maximum value among the global similarity values of each video frame, then determine this video frame corresponding to the maximum value as the standard frame.

[0131] Step S304: If there are multiple video frames corresponding to the maximum value among the global similarity values of each video frame, then determine the video frame located in the middle position in the time series among the multiple video frames as the standard frame.

[0132] Specifically, according to the calculation process in step S302, if it is determined that there are multiple video frames corresponding to the maximum value among the global similarity values of each video frame, then determine the video frame located in the middle position in the time series among the multiple video frames as the standard frame to ensure the coherence of time.

[0133] Among them, when the number of multiple video frames is odd, take the middle frame as the standard frame; when the number of multiple video frames is even, take the frame with the smaller serial number among the two middle frames in the time series as the standard frame.

[0134] The process of determining the standard frame of the sub - fragment to be processed provided by the embodiments of the present application is as follows: by obtaining at least one normalized histogram of each video frame in the sub - fragment to be processed, calculating the global similarity value of each video frame in the sub - fragment to be processed according to the at least one normalized histogram. If there is only one video frame corresponding to the maximum value among the global similarity values of each video frame, then this video frame is determined as the standard frame; if there are multiple video frames corresponding to the maximum value among the global similarity values of each video frame, then the video frame located in the middle position in the time series among the multiple video frames is determined as the standard frame. Among them, in the process of determining the standard frame, by selecting the video frame with the highest similarity to other video frames in the sub - fragment to be processed through global similarity as the standard frame, it can make the optimization process based on this standard frame more accurate, thereby improving the accuracy of video processing and further enhancing the user's video viewing experience. Among them, in the process of determining the global similarity value of each video frame, based on at least one normalized histogram, it can improve the comprehensiveness and accuracy of global similarity calculation, thereby improving the accuracy of video processing and further enhancing the user's video viewing experience. Based on the above description, the process of determining the standard frame of the sub - fragment to be processed provided by the embodiments of the present application can improve the user's video viewing experience.

[0135] Figure 4 Schematic flow of the method for removing breathing flicker from video frames provided by the present application Figure 4 , such as Figure 4 shown, based on the embodiments of Figure 1 or Figure 2 or Figure 3 On the basis of the embodiment, the process of optimizing the non - standard frames of the sub - fragment to be processed according to the standard frame to obtain the sub - fragment of the video to be processed after breathing flicker removal is described in detail. The method includes:

[0136] Step S401: Obtain at least one normalized histogram of each video frame in the sub - fragment to be processed.

[0137] Specifically, at least one normalized histogram of each video frame in the sub - fragment to be processed can be obtained. The specific description of this step can refer to the description in step S301 and will not be elaborated here.

[0138] Step S402: Calculate the normalized histogram cumulative distribution function value of each video frame in the sub - fragment to be processed according to the at least one normalized histogram.

[0139] Specifically, according to the at least one normalized histogram obtained in step S401, the normalized histogram cumulative distribution function value of each video frame in the sub - fragment to be processed can be calculated. The formula for calculating the normalized histogram cumulative distribution function value of the standard frame in the sub - fragment to be processed is as follows:

[0140]

[0141] Among them, CDF ref (C) represents the normalized histogram cumulative distribution function value of the standard frame, C represents the set composed of the normalized histograms of the video frames, where C ∈ {V: luminance, H: hue, S: saturation}, and L represents the number of pixel values in the normalized histogram, for example, 256.

[0142] Among them, the formula for calculating the normalized histogram cumulative distribution function value of the non-standard frames in the sub-segment to be processed is as follows:

[0143]

[0144] Among them, CDF i (C) represents the normalized histogram cumulative distribution function value of the non-standard frame.

[0145] Among them, the normalized histogram cumulative distribution function value of the non-standard frames in the sub-segment to be processed described above and the calculated normalized histogram cumulative distribution function value of the non-standard frames in the sub-segment to be processed are the cumulative distribution function values corresponding to the luminance color channel.

[0146] Among them, comprehensively considering the normalized histograms of each color channel described in step S401, for example, in the process of calculating the normalized histogram cumulative distribution function value of the non-standard frames in the sub-segment to be processed and the calculated normalized histogram cumulative distribution function value of the non-standard frames in the sub-segment to be processed, the cumulative distribution function values corresponding to the normalized histograms of each color channel are calculated respectively.

[0147] Step S403: According to the normalized histogram cumulative distribution function value of the standard frame, perform a mapping process on the normalized histogram cumulative distribution function value of each non-standard frame to obtain the sub-segment of the video to be processed after removing breathing and flickering.

[0148] Specifically, according to the normalized histogram cumulative distribution function value of the standard frame obtained in step S402, a mapping process can be performed on the normalized histogram cumulative distribution function value of each non-standard frame to obtain the sub-segment of the video to be processed after removing breathing and flickering. Among them, the calculation formula in the mapping process is as follows:

[0149]

[0150] Among them, y is the mapping value in the mapping process. Specifically, the mapping value is to find the pixel value closest to CDF ref (H) for each pixel value x through the standard frame CDF i (H).

[0151] Optionally, during the process of calculating the mapping value, linear interpolation can be used to improve the mapping accuracy and avoid the staircase effect.

[0152] Among them, the above-described mapping process corresponds to the luminance color channel. Specifically, for the hue color channel and the saturation color channel, the above-described mapping process can also be referred to, which will not be elaborated here.

[0153] Optionally, during the mapping process corresponding to the hue color channel, edge continuity processing can be performed first.

[0154] Specifically, after each color channel undergoes the above-described mapping process, they can be combined to obtain the processed HSV image, and then the HSV image is converted to the RGB space to obtain the processed video clip after breathing and flickering removal.

[0155] In the process of optimizing the non-standard frames of the subclip to be processed according to the standard frame provided by the embodiment of the present application to obtain the processed video clip after breathing and flickering removal, by obtaining at least one normalized histogram of each video frame in the subclip to be processed, calculating the normalized histogram cumulative distribution function values of each video frame in the subclip to be processed according to at least one normalized histogram, and performing mapping processing on the normalized histogram cumulative distribution function value of each non-standard frame according to the normalized histogram cumulative distribution function value of the standard frame to obtain the processed video clip after breathing and flickering removal, wherein, performing mapping processing on the normalized histogram cumulative distribution function value of each non-standard frame according to the normalized histogram cumulative distribution function value of the standard frame can accurately and efficiently perform breathing and flickering removal processing, thereby improving the user's video viewing experience. Among them, during the mapping process, based on at least one normalized histogram, the comprehensiveness and accuracy of breathing and flickering removal processing can be improved, further improving the user's video viewing experience. Combining the above description, the process of optimizing the non-standard frames of the subclip to be processed according to the standard frame provided by the embodiment of the present application to obtain the processed video clip after breathing and flickering removal can improve the user's video viewing experience.

[0156] Figure 5 It is a flowchart showing the method for removing breathing and flickering from the video frames provided by the present application Figure 5 , as Figure 5 shown. In this embodiment, based on the Figure 1 or Figure 2 or Figure 3 or Figure 4 embodiment, the process of another method for removing breathing and flickering from the video frames is described in detail. The method includes:

[0157] Step S501, obtain the subclip to be processed after de-flickering processing.

[0158] Specifically, the specific process of this step can refer to the description in step S101, which will not be elaborated here.

[0159] Step S502: According to the local space information capacity of the to-be-processed video clip after de-flickering processing, perform similarity division processing on the to-be-processed video clip after de-flickering processing to obtain at least one to-be-processed sub-clip.

[0160] Specifically, the specific process of this step can refer to the description in step S102, which will not be elaborated here.

[0161] Step S503: Determine the standard frame of the to-be-processed sub-clip, and optimize the non-standard frames of the to-be-processed sub-clip according to the standard frame to obtain the to-be-processed sub-clip after optimization processing.

[0162] Specifically, the specific process of this step can refer to the description in step S103, which will not be elaborated here.

[0163] Step S504: Calculate the luminance histogram cumulative distribution function values of the standard frames of each to-be-processed sub-clip after optimization processing, and determine the luminance mapping relationship between adjacent to-be-processed sub-clips after optimization processing based on the luminance histogram cumulative distribution function values.

[0164] Specifically, after completing the optimization processing of the non-standard frames of each to-be-processed sub-clip according to the standard frame, the luminance histogram of the standard frame of each to-be-processed sub-clip after optimization processing can be obtained first, and then the luminance histogram cumulative distribution function values of the standard frames of each to-be-processed sub-clip after optimization processing can be calculated according to the luminance histogram of the standard frame of each to-be-processed sub-clip after optimization processing.

[0165] Specifically, the formula for the luminance histogram of the standard frame of the to-be-processed sub-clip after optimization processing is as follows:

[0166]

[0167] where n s (k) is the number of pixels with luminance value k in the standard frame F s of the to-be-processed sub-clip after optimization processing, and N s is the total number of pixels of the standard frame F s of the to-be-processed sub-clip after optimization processing.

[0168] Specifically, the formula for the luminance histogram cumulative distribution function value of the standard frame of the to-be-processed sub-clip after optimization processing is as follows:

[0169]

[0170] Among them, the value C of the cumulative distribution function of the luminance histogram of the standard frame of the sub - fragment to be processed after optimization s (k) reflects the cumulative proportion of pixels with luminance values less than or equal to k in the standard frame of the sub - fragment to be processed after optimization, providing key data for determining the luminance mapping relationship between adjacent sub - fragments subsequently.

[0171] Specifically, the formula for the luminance mapping relationship between every two adjacent sub - fragments to be processed after optimization is as follows:

[0172]

[0173] Among them, M(k) represents the luminance mapping relationship between two adjacent sub - fragments to be processed after optimization, and this luminance mapping relationship is determined by minimizing the difference in the cumulative distribution functions of adjacent standard frames. Among them, represents the value of the cumulative distribution function of the luminance histogram of the standard frame of the i S - th sub - fragment to be processed after optimization, represents the value of the cumulative distribution function of the luminance histogram of the standard frame of the i+1 (S + 1)-th sub - fragment to be processed after optimization.

[0174] Specifically, for two adjacent sub - fragments to be processed after optimization, S i and S i+1 , the following two - direction luminance mapping relationships are defined: First, the forward mapping function M i→i+1 (k): Maps the luminance value of the i - th sub - fragment to be processed after optimization to the luminance space of the (i + 1)-th sub - fragment to be processed after optimization, and its determination process satisfies , and the formula is:

[0175]

[0176] Among them: C i (k) is the value of the cumulative distribution function of the luminance histogram of the standard frame of the sub - fragment to be processed after optimization, S i , is the inverse function of the value of the cumulative distribution function of the luminance histogram of the standard frame of the sub - fragment to be processed after optimization, S i+1 (S + 1). This formula converts the pixel with luminance value k in the i - th sub - fragment to be processed after optimization into a probability value (cumulative proportion) through its cumulative distribution function C i (k), and then uses the inverse function of the cumulative distribution function of the (i + 1)-th sub - fragment to be processed after optimization to map this probability value back to the luminance space of the (i + 1)-th sub - fragment to be processed after optimization, obtaining Mi→i+1 (k).

[0177] Second, the backward mapping function : Maps the luminance value of the (i + 1)-th sub-fragment to the luminance space of the -th sub-fragment, and its determination process satisfies , and the formula is:

[0178]

[0179] Among them, the specific description of this formula can refer to the description of the forward mapping function, which will not be elaborated here.

[0180] Step S505: Set an overlapping area between adjacent processed sub-fragments to be processed, and perform luminance interpolation adjustment on each video frame in the overlapping area based on the luminance mapping relationship to obtain two sets of overlapping areas with luminance interpolation adjustment.

[0181] Among them, the overlapping area includes all boundary frames of adjacent processed sub-fragments to be processed.

[0182] Specifically, to ensure a smooth transition of the regional energy distribution and gray complexity between adjacent processed sub-fragments to be processed S i and S i+1 , an overlapping area O can be set between them. The overlapping area O includes the latter part of the frames of the previous processed sub-fragment to be processed S i , that is, the posterior boundary frames, and the former part of the frames of the latter processed sub-fragment to be processed S i+1 , that is, the anterior boundary frames, where the latter part of the frames and the former part of the frames include multiple video frames.

[0183] Specifically, the process of performing luminance interpolation adjustment on each video frame in the overlapping area based on the luminance mapping relationship to obtain two sets of overlapping areas with luminance interpolation adjustment can be calculated by a linear interpolation algorithm, as follows:

[0184] Specifically, for any frame in the overlapping area, two adjusted luminance values are generated based on the bidirectional luminance mapping relationship.

[0185] Among them, the adjustment value y1 based on the forward mapping.

[0186]

[0187] Specifically, for one of the overlapping areas with luminance interpolation adjustment, each video frame in the overlapping area gradually transitions to the processed sub-fragment to be processed S i+1The brightness feature, where x represents the current brightness value of each video frame in the overlapping area before brightness interpolation adjustment, that is, the original brightness value possessed by the video frame itself. α is the interpolation coefficient, and its value range is [0, 1]. By adjusting α, the degree of brightness adjustment can be controlled, and the brightness mutation that may occur at the boundary of adjacent sub - segments can be reduced.

[0188] Among them, the adjustment value y2 based on backward mapping.

[0189]

[0190] Specifically, for one overlapping area after brightness interpolation adjustment, each video frame in the overlapping area gradually transitions to the brightness feature of sub - segment S i .

[0191] Among them, for each video frame in the overlapping area, there are two adjusted brightness values. Among each group of overlapping areas after brightness interpolation adjustment, the adjusted brightness values corresponding to the same video frame are different. This is because each overlapping area frame needs to retain two adjustment results simultaneously for subsequent weighted fusion through the transparency channel to achieve smooth transition. For example, for the post - boundary frame of sub - segment S i , α can linearly increase from 0 to 1; for the pre - boundary frame of sub - segment S i+1 , α can linearly decrease from 1 to 0, so as to ensure that the video picture is more coherent in terms of brightness.

[0192] Step S506: Generate the value of the transparency channel corresponding to the overlapping area, and perform weighted fusion processing on the pixel brightness values of each video frame in the two groups of overlapping areas after brightness interpolation adjustment to obtain the video clip to be processed after removing breathing flicker.

[0193] Specifically, after setting the overlapping area O, for the overlapping area O, the corresponding value T of the transparency channel can be generated. Let the number of frames in the overlapping area O be N O , the number of frames in the overlapping area O in the previous video clip to be processed S i after brightness interpolation adjustment is , and the number of frames in the overlapping area O in the next video clip to be processed S i+1 after brightness interpolation adjustment is , and . Among them, the value T of the transparency channel linearly decreases from 1 to 0 in the overlapping area of the previous video clip to be processed S i after brightness interpolation adjustment, and linearly increases from 0 to 1 in the overlapping area of the next video clip to be processed S i+1 after brightness interpolation adjustment. Specifically, for the nth frame (n = 1, ⋯, N O ) in the overlapping area O, the value T(n) of the transparency channel can be calculated by the following formula:

[0194]

[0195] Moreover, the value T of the transparency channel is generated based on the local spatial information sub-capacity feature of the sub-fragment to be processed after brightness interpolation adjustment. For example, it can be adjusted according to information such as the local stability index (LSIC) value of the pixel block in adjacent sub-fragments to ensure smooth transition of the regional energy distribution and gray complexity of adjacent sub-fragments to be processed after brightness interpolation adjustment, and avoid obvious visual breaks.

[0196] Specifically, according to the generated value T(n) of the transparency channel, the pixel brightness values of each video frame in the overlapping area of the two groups of sub-fragments to be processed after brightness interpolation adjustment can be weighted and fused to obtain the sub-fragment of the video to be processed after removing breathing flicker. Specifically, the formula in the process of weighted fusion processing is as follows:

[0197]

[0198] where I(n) is the fused brightness value, I i (n) and I i+1 (n) are the pixel brightness values of the nth video frame in the overlapping area of the two groups of sub-fragments to be processed after brightness interpolation adjustment. Among them, I i (n) is the brightness value of the pixel of the nth frame in the overlapping area O after adjustment according to the standard frame of the sub-fragment to be processed S i after optimization processing in the sub-fragment to be processed S i+1 , and I i+1 (n) is the brightness value of the pixel of the nth frame in the overlapping area O after adjustment according to the standard frame of the sub-fragment to be processed S i+1 after optimization processing in the sub-fragment to be processed S i . Specifically, after such weighted fusion processing, the sub-fragment of the video to be processed after removing breathing flicker is finally obtained, and the video sub-fragment is more natural and smooth in terms of brightness and picture transition.

[0199] The further processing procedure after optimizing the non-standard frames of the sub-fragments to be processed according to the standard frames provided by the embodiments of the present application includes calculating the luminance histogram cumulative distribution function values of the standard frames of each sub-fragment to be processed after optimization, determining the luminance mapping relationship between adjacent sub-fragments to be processed after optimization based on the luminance histogram cumulative distribution function values, and performing luminance interpolation adjustment on the boundary frames of adjacent sub-fragments to be processed after optimization based on the luminance mapping relationship to obtain the sub-fragments to be processed after luminance interpolation adjustment. An overlapping area is set between adjacent sub-fragments to be processed after luminance interpolation adjustment, and the values of the transparency channel corresponding to the overlapping area are generated. The pixel luminance values of each video frame in the overlapping area are weighted and fused according to the values of the transparency channel to obtain the video segment to be processed after removing breathing flicker. Among them, the luminance mapping relationship is accurately determined and the luminance interpolation adjustment is performed through mathematical formulas, and the weighted fusion is performed using the values of the transparency channel, which can accurately and efficiently achieve smooth transition between adjacent sub-fragments, reduce luminance mutation and visual tomogram, and further improve the effect of removing breathing flicker. At the same time, considering the local spatial information sub-capacity characteristics to generate the values of the transparency channel ensures the coherence and naturalness of the video picture in terms of regional energy distribution and gray complexity, thus significantly improving the user's video viewing experience. Based on the above description, this subsequent processing procedure provided by the embodiments of the present application can further improve the quality of video processing and bring a better viewing experience to users.

[0200] Figure 6 FIG. is a schematic structural diagram of a device for removing breathing flicker from a video picture provided by the present application, as Figure 6 shown, the device 60 for removing breathing flicker from a video picture provided in this embodiment includes:

[0201] An acquisition module 601, configured to acquire the video segment to be processed after de-flickering processing; wherein, the video segment to be processed is a video segment in the same video scene; the video segment to be processed after de-flickering processing is obtained by correcting the exposure abnormal frames; the exposure abnormal frames are determined from multiple video frames of the video segment to be processed according to the local spatial information capacity; the local spatial information capacity is used to characterize the regional energy distribution of the video frames;

[0202] A first processing module 602, configured to perform similarity division processing on the video segment to be processed after de-flickering processing according to the local spatial information capacity of the video segment to be processed after de-flickering processing, to obtain at least one sub-fragment to be processed;

[0203] A second processing module 603, configured to determine the standard frames of the sub-fragments to be processed, and optimize the non-standard frames of the sub-fragments to be processed according to the standard frames, to obtain the video segment to be processed after removing breathing flicker.

[0204] In a possible embodiment, the local space information capacity includes at least one local space information sub-capacity; wherein, the local space information sub-capacity is used to characterize the regional energy distribution of pixel blocks in a video frame; the first processing module 602 is specifically configured to determine the similarity between every two video frames in the to-be-processed video segment after de-flickering processing according to at least one local space information sub-capacity included in the local space information capacity; and perform similarity division processing on the to-be-processed video segment after de-flickering processing according to the similarity between every two video frames, to obtain at least one to-be-processed sub-segment.

[0205] In a possible embodiment, the first processing module 602 is further specifically configured to perform feature extraction processing on at least one local space information sub-capacity included in the local space information capacity, to obtain the local space information sub-capacity feature vectors of each video frame; and perform cosine similarity calculation processing on the local space information sub-capacity feature vectors of each video frame, to obtain the similarity between every two video frames.

[0206] In a possible implementation manner, the first processing module 602 is further specifically configured to perform feature extraction processing on at least one local space information sub-capacity included in the local space information capacity, to obtain the local space information sub-capacity feature vectors of each video frame; obtain the information entropy feature vectors of each video frame, and perform splicing processing on the local space information sub-capacity feature vectors and the information entropy feature vectors, to obtain the integrated information feature vectors of each video frame; and perform cosine similarity calculation processing on the integrated information feature vectors of each video frame, to obtain the similarity between every two video frames.

[0207] In a possible embodiment, the first processing module 602 is further specifically configured to determine at least one video segmentation point based on the dynamic programming technique according to the similarity between every two video frames; wherein, the video segmentation point is used to divide two to-be-processed segments; and perform segmentation processing on the to-be-processed video segment according to at least one video segmentation point, to obtain at least one to-be-processed sub-segment.

[0208] In a possible embodiment, the second processing module 603 is specifically configured to obtain at least one normalized histogram of each video frame in the to-be-processed sub-segment; calculate the global similarity value of each video frame in the to-be-processed sub-segment according to at least one normalized histogram; wherein, the global similarity value is the sum value of the similarity values between a video frame and other video frames; if there is only one video frame corresponding to the maximum value among the global similarity values of each video frame, then determine this video frame as the standard frame; if there are multiple video frames corresponding to the maximum value among the global similarity values of each video frame, then determine the video frame at the middle position in the time series among the multiple video frames as the standard frame.

[0209] In a possible embodiment, the second processing module 603 is further specifically configured to obtain at least one normalized histogram of each video frame in the sub - segment to be processed; calculate the normalized histogram cumulative distribution function values of each video frame in the sub - segment to be processed according to the at least one normalized histogram; perform a mapping process on the normalized histogram cumulative distribution function value of each non - standard frame according to the normalized histogram cumulative distribution function value of the standard frame, so as to obtain the sub - segment of the video to be processed after removing breathing and flickering.

[0210] In a possible embodiment, the second processing module 603 is further configured to calculate the luminance histogram cumulative distribution function values of the standard frames of each sub - segment of the video to be processed after optimization processing, and determine the luminance mapping relationship between adjacent sub - segments of the video to be processed after optimization processing based on the luminance histogram cumulative distribution function values; set an overlapping area between adjacent sub - segments of the video to be processed after optimization processing, and perform luminance interpolation adjustment on each video frame in the overlapping area based on the luminance mapping relationship to obtain two sets of overlapping areas after luminance interpolation adjustment; wherein, the overlapping area includes all boundary frames of adjacent sub - segments of the video to be processed after optimization processing; generate the values of the transparency channel corresponding to the overlapping area, and perform weighted fusion processing on the pixel luminance values of each video frame in the two sets of overlapping areas after luminance interpolation adjustment according to the values of the transparency channel, so as to obtain the sub - segment of the video to be processed after removing breathing and flickering.

[0211] The apparatus for removing breathing and flickering of video images provided in this embodiment can execute the method provided in the above - mentioned method embodiment, and its implementation principle and technical effects are similar, so details are not described here in this embodiment.

[0212] Figure 7 This is a schematic structural diagram of the electronic device provided in this application. As Figure 7 shown, the electronic device 70 provided in this embodiment includes: at least one processor 701 and a memory 702. Optionally, the electronic device 70 further includes a communication component 703. Among them, the processor 701, the memory 702, and the communication component 703 are connected through a bus 704.

[0213] In a specific implementation process, at least one processor 701 executes the computer - executable instructions stored in the memory 702, so that at least one processor 701 executes the above - mentioned method.

[0214] The specific implementation process of the processor 701 can be referred to in the above - mentioned method embodiment, and its implementation principle and technical effects are similar, so details are not described here again in this embodiment.

[0215] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU for short), or other general-purpose processors, digital signal processors (DSP for short), application specific integrated circuits (ASIC for short), etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the method disclosed in combination with the invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor.

[0216] The memory may include a high-speed random access memory (RAM), and may also include non-volatile memory (NVM), such as at least one disk memory.

[0217] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, the buses in the drawings of this application are not limited to only one bus or one type of bus.

[0218] This application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0219] This application also provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the processor executes the computer-executable instructions, the above method is implemented.

[0220] The above-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, a magnetic disk or an optical disk. The readable storage medium can be any available medium accessible by a general-purpose or special-purpose computer.

[0221] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an Application Specific Integrated Circuits (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in a device.

[0222] The division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Additionally, the couplings or direct couplings or communication connections shown or discussed between each other can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0223] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0224] Furthermore, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0225] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0226] Those of ordinary skill in the art will understand that all or part of the steps to implement the above method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps including the above method embodiments; and the aforementioned storage medium includes: various media such as ROM, RAM, magnetic disks, or optical discs that can store program codes.

[0227] Finally, it should be noted that: After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily conceive of other embodiments of the present invention. The present invention is intended to cover any variations, uses, or adaptations of the present invention, which follow the general principles of the present invention and include common general knowledge or conventional technical means in the technical field not disclosed by the present invention. It is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A method for removing breathing flicker from a film picture, characterized in that, Including: Obtaining a to-be-processed video clip after de-flickering processing; wherein, the to-be-processed video clip is a clip of video frames under the same video scene; the to-be-processed video clip after de-flickering processing is obtained by correcting an exposure abnormal frame; the exposure abnormal frame is determined from multiple video frames of the to-be-processed video clip according to the local space information capacity; the local space information capacity is used to characterize the regional energy distribution of the video frame; Performing a similarity division process on the to-be-processed video clip after de-flickering processing according to the local space information capacity of the to-be-processed video clip after de-flickering processing to obtain at least one to-be-processed sub-clip; Determining a standard frame of the to-be-processed sub-clip and optimizing non-standard frames of the to-be-processed sub-clip according to the standard frame to obtain a to-be-processed video clip after breathing flicker removal processing.

2. The method according to claim 1, wherein The local space information capacity includes at least one local space information sub-capacity; wherein, the local space information sub-capacity is used to characterize the regional energy distribution of pixel blocks in the video frame; Performing a similarity division process on the to-be-processed video clip after de-flickering processing according to the local space information capacity of the to-be-processed video clip after de-flickering processing to obtain at least one to-be-processed sub-clip, including: Determining the similarity between every two video frames in the to-be-processed video clip after de-flickering processing according to at least one of the local space information sub-capacities included in the local space information capacity; Performing a similarity division process on the to-be-processed video clip after de-flickering processing according to the similarity between every two video frames to obtain at least one to-be-processed sub-clip.

3. The method according to claim 2, wherein Determining the similarity between every two video frames in the to-be-processed video clip after de-flickering processing according to at least one of the local space information sub-capacities included in the local space information capacity, including: Performing feature extraction processing on at least one of the local space information sub-capacities included in the local space information capacity to obtain local space information sub-capacity feature vectors of each video frame; Performing a cosine similarity calculation process on the local space information sub-capacity feature vectors of each video frame to obtain the similarity between every two video frames.

4. The method according to claim 2, wherein Determining the similarity between every two video frames in the to-be-processed video clip after de-flickering processing according to at least one of the local space information sub-capacities included in the local space information capacity, including: Performing feature extraction processing on at least one of the local space information sub-capacities included in the local space information capacity to obtain local space information sub-capacity feature vectors of each video frame; Obtaining information entropy feature vectors of each video frame and performing a splicing process on the local space information sub-capacity feature vectors and the information entropy feature vectors to obtain integrated information feature vectors of each video frame; Performing a cosine similarity calculation process on the integrated information feature vectors of each video frame to obtain the similarity between every two video frames.

5. The method according to claim 2, wherein Performing a similarity division process on the de - flickered to - be - processed video clip according to the similarity between every two video frames, to obtain at least one to - be - processed sub - clip, including: Based on the similarity between every two video frames and using dynamic programming techniques, determining at least one video segmentation point; wherein, the video segmentation point is used to divide two of the to - be - processed segments; Performing a segmentation process on the to - be - processed video clip according to the at least one video segmentation point, to obtain the at least one to - be - processed sub - clip.

6. The method according to claim 1, characterized in that Determining the standard frame of the to - be - processed sub - clip, including: Obtaining at least one normalized histogram of each video frame in the to - be - processed sub - clip; According to the at least one normalized histogram, calculating the global similarity value of each video frame in the to - be - processed sub - clip; wherein, the global similarity value is the sum of the similarity values between a video frame and other video frames; If there is one video frame corresponding to the maximum value among the global similarity values of each video frame, determining this video frame as the standard frame; If there are multiple video frames corresponding to the maximum value among the global similarity values of each video frame, determining the video frame at the middle position in the time series among the multiple video frames as the standard frame.

7. The method according to claim 1, characterized in that, Performing an optimization process on the non - standard frames of the to - be - processed sub - clip according to the standard frame, to obtain the to - be - processed video clip after breathing flicker removal, including: Obtaining at least one normalized histogram of each video frame in the to - be - processed sub - clip; According to the at least one normalized histogram, calculating the normalized histogram cumulative distribution function value of each video frame in the to - be - processed sub - clip; According to the normalized histogram cumulative distribution function value of the standard frame, performing a mapping process on the normalized histogram cumulative distribution function value of each non - standard frame, to obtain the to - be - processed video clip after breathing flicker removal.

8. The method according to any one of claims 1 to 7, characterized in that, After performing the optimization process on the non - standard frames of the to - be - processed sub - clip according to the standard frame, the method further includes: Calculating the luminance histogram cumulative distribution function value of the standard frame of each optimized to - be - processed sub - clip, and based on the luminance histogram cumulative distribution function value, determining the luminance mapping relationship between adjacent optimized to - be - processed sub - clips; Setting an overlap area between adjacent optimized to - be - processed sub - clips, and based on the luminance mapping relationship, performing luminance interpolation adjustment on each video frame in the overlap area, to obtain two sets of video frames in the overlap area after luminance interpolation adjustment; wherein, the overlap area includes all boundary frames of adjacent optimized to - be - processed sub - clips; Generating the value of the transparency channel corresponding to the overlap area, and according to the value of the transparency channel, performing weighted fusion processing on the pixel luminance values of each video frame in the two sets of video frames in the overlap area after luminance interpolation adjustment, to obtain the to - be - processed video clip after breathing flicker removal.

9. A device for removing breathing flicker from a film picture, characterized in that, Including: An acquisition module for acquiring a to-be-processed video clip after de-flickering processing; wherein, the to-be-processed video clip is a clip of video frames in the same video scene; the to-be-processed video clip after de-flickering processing is obtained by correcting an exposure abnormal frame; the exposure abnormal frame is determined from multiple video frames of the to-be-processed video clip according to the local space information capacity; the local space information capacity is used to characterize the regional energy distribution of the video frame. A first processing module for performing a similarity division process on the to-be-processed video clip after de-flickering processing according to the local space information capacity of the to-be-processed video clip after de-flickering processing, to obtain at least one to-be-processed sub-clip. A second processing module for determining a standard frame of the to-be-processed sub-clip and optimizing non-standard frames of the to-be-processed sub-clip according to the standard frame, to obtain a to-be-processed video clip after breathing flicker removal processing.

10. An electronic device, characterized in that, Comprising: A memory, a processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor executes the method according to any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, Computer-executable instructions are stored in the computer-readable storage medium, and when the computer-executable instructions are executed by a processor, they are used to implement the method according to any one of claims 1-8.

12. A computer program product, characterized in that, Comprising a computer program, which when executed by a processor implements the method according to any one of claims 1-8.

Images