CN107483950A - Picture parallel encoding method and system - Google Patents

Picture parallel encoding method and system Download PDF

Info

Publication number
CN107483950A
CN107483950A CN201610399069.5A CN201610399069A CN107483950A CN 107483950 A CN107483950 A CN 107483950A CN 201610399069 A CN201610399069 A CN 201610399069A CN 107483950 A CN107483950 A CN 107483950A
Authority
CN
China
Prior art keywords
coding unit
coding
picture
row
unit row
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201610399069.5A
Other languages
Chinese (zh)
Inventor
范英明
孙俊
郭宗明
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Peking University
Peking University Founder Group Co Ltd
Beijing Founder Electronics Co Ltd
Original Assignee
Peking University
Peking University Founder Group Co Ltd
Beijing Founder Electronics Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Peking University, Peking University Founder Group Co Ltd, Beijing Founder Electronics Co Ltd filed Critical Peking University
Priority to CN201610399069.5A priority Critical patent/CN107483950A/en
Publication of CN107483950A publication Critical patent/CN107483950A/en
Pending legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/42—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
    • H04N19/436—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation using parallelised computational arrangements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/156—Availability of hardware or computational resources, e.g. encoding based on power-saving criteria
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/172—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a picture, frame or field

Abstract

The invention provides a kind of picture parallel encoding method and a kind of picture parallel encoder system, wherein, the picture parallel encoding method includes：Picture to be encoded is divided into multiple coding regions, each coding region corresponds to a coding unit row in image encoder respectively, and each coding unit row includes multiple coding units；Parallel encoding of multiple coding unit rows to the multiple coding region is started according to default startup relation successively.By technical scheme, coding of graphics speed can be effectively improved, while higher compression efficiency is ensured so as to lift Consumer's Experience.

Description

Picture parallel encoding method and system
Technical field
The present invention relates to technical field of information processing, in particular to a kind of picture parallel encoding side Method and a kind of picture parallel encoder system.
Background technology
With the development of information technology, the status of picture becomes more and more important in information displaying, one The volume of a width picture is the several times of volume shared by wherein all words in webpage, it is clear that picture volume Bigger, the time that user loads webpage is longer.
HEVC (High Efficiency Video Coding) is newest video encoding standard, its phase For previous generation standards H.264/AVC (Advanced Video Coding, advanced video coding) and Speech, under equal objective quality, its compression ratio can double, and can adapt to be widely applied needs Ask.And HEVC can be equally applied in coding of graphics, and 25% or so compression can be brought to imitate Rate is lifted, and still, encoder complexity is also accordingly lifted so that is lifted while compression efficiency is kept Code rate becomes particularly important.
Therefore, how while higher compression efficiency is ensured, coding of graphics speed is effectively improved Rate, so as to lift Consumer's Experience, turn into technical problem urgently to be resolved hurrily.
The content of the invention
The present invention is based on above-mentioned technical problem, it is proposed that a kind of new technical scheme, it is possible to achieve Coding of graphics parallelization based on HEVC standard, while higher compression efficiency is ensured, to have Effect ground improves coding of graphics speed, so as to lift Consumer's Experience.
In view of this, first aspect of the present invention it is proposed a kind of picture parallel encoding method, bag Include：Picture to be encoded is divided into multiple coding regions, each coding region corresponds to picture respectively A coding unit row in encoder, and each coding unit row includes multiple coding lists Member；Multiple coding unit rows are started to the multiple coding region according to default startup relation successively Parallel encoding.
In the technical scheme, picture to be encoded is divided into by multiple code areas by image encoder Domain, and it is right respectively by the different coding unit rows comprising multiple coding units of the image encoder Corresponding coding region is encoded, moreover, closed according to default start between multiple coding unit rows System starts coding work and carries out parallel encoding successively, so, it is possible to achieve coding of graphics parallelization, have Effect ground improves coding of graphics speed, so as to lift Consumer's Experience.
In the above-mentioned technical solutions, it is preferable that also include：Determine to be used to perform according to CPU core number The destination number of the thread of the encoding tasks of multiple coding unit rows, wherein, the destination number More than the CPU core number, and the corresponding volume for performing a coding unit row of each thread Code task.
In the technical scheme, specifically, when each coding unit row performs encoding tasks, journey is encoded For sort run on corresponding thread, i.e., each thread performs the encoding tasks of a coding unit row, and The destination number of the thread needed can according to the CPU of computer (Central Processing Unit, Central processing unit) check figure determination, wherein, the destination number of thread is more than CPU core number, Ke Yishi Its multiple, than the thread that if any 4 cores, then can start 1.5 times of CPU core number, i.e. destination number For 6, CPU core number at least two, i.e., by multicore architecture, coding of graphics can be better achieved Parallelization, and then improve coding of graphics speed.
In any of the above-described technical scheme, it is preferable that the default startup relation of the basis starts more successively The individual coding unit row specifically includes to the parallel encoding of the multiple coding region：Detect previous Whether second coding unit in the individual coding unit row is completed to encode；It is previous detecting When second in the individual coding unit row coding unit completes coding, start adjacent latter The encoding tasks of first coding unit in the individual coding unit row；And upon actuation one The coding of other coding units in addition to first coding unit in the individual coding unit row is appointed Before business, coding list adjacent with other described coding units in the previous coding unit row is judged Whether member is completed to encode.
In the technical scheme, treated starting multiple coding unit rows successively according to default startup relation When multiple coding regions of encoded picture carry out parallel encoding, the first coding of the latter coding unit row Whether the startup of unit needs to complete to encode according to the second coding unit of adjacent previous coding unit row Depending on, i.e., the entropy code state of the first coding unit of current coded unit row is needed from adjacent previous Second coding unit of individual coding unit row obtains, if the second coding unit of previous coding unit row The first coding unit that completion coding can just start the latter coding unit row is encoded, with such Push away, to realize that parallelization encodes；Moreover, upon actuation in a coding unit row from second coding Before the encoding tasks for other coding units that unit starts, except the previous volume in this coding unit row It is outer that code unit completes coding, it is also necessary to judge in adjacent previous coding unit row with current coded unit Whether adjacent coding unit, which is completed, encodes, if so, can then start the coding of the current coded unit Task, otherwise wait for temporarily not starting coding work, to realize that parallelization encodes effectively, in an orderly manner.
In any of the above-described technical scheme, it is preferable that when the destination number is less than multiple codings The quantity of cell row, and each thread has correspondingly performed the coding times of a coding unit row During business, the thread is discharged, for performing the encoding tasks of other coding unit rows.
In the technical scheme, thread may be reused, and particularly be determined according to CPU core number Thread destination number be less than image encoder in coding unit row quantity when, to make full use of Existing resource completes picture parallel encoding task.
In any of the above-described technical scheme, it is preferable that described more the picture to be encoded is divided into Before individual coding region, in addition to：Treated by the image encoder based on HEVC standard to described Encoded picture carries out precoding processing.
In the technical scheme, in order to ensure picture compression efficiency, it is necessary to treating encoded picture draw / preceding, the picture to be encoded is carried out at the precoding based on HEVC standard by image encoder Reason, to be simplified to code stream header file so that picture is more suitable for parallel encoding, in this way, The coding of graphics parallelization based on HEVC standard is realized, to ensure the same of higher compression efficiency When, coding of graphics speed is effectively improved, so as to lift Consumer's Experience.
In addition, after all threads are finished, i.e., all coding unit rows are to corresponding code area When coding is completed in domain, then each coding region after the completion of integration encodes, the picture formed after coding, Terminate picture parallel encoding task.
The second aspect of the present invention, it is proposed that a kind of picture parallel encoder system, including：Divide mould Block, for picture to be encoded to be divided into multiple coding regions, each coding region corresponds to respectively A coding unit row in image encoder, and each coding unit row includes multiple codings Unit；Control module, for starting multiple coding units successively according to default startup relation control The parallel encoding for the multiple coding region that row divides to obtain to the division module.
In the technical scheme, picture to be encoded is divided into by multiple code areas by image encoder Domain, and it is right respectively by the different coding unit rows comprising multiple coding units of the image encoder Corresponding coding region is encoded, moreover, closed according to default start between multiple coding unit rows System starts coding work and carries out parallel encoding successively, so, it is possible to achieve coding of graphics parallelization, have Effect ground improves coding of graphics speed, so as to lift Consumer's Experience.
In the above-mentioned technical solutions, it is preferable that also include：Determining module, for according to CPU core Number determines the destination number for being used to perform the thread of the encoding tasks of multiple coding unit rows, its In, the destination number is more than the CPU core number, and each one institute of the corresponding execution of the thread State the encoding tasks of coding unit row.
In the technical scheme, specifically, when each coding unit row performs encoding tasks, journey is encoded For sort run on corresponding thread, i.e., each thread performs the encoding tasks of a coding unit row, and The destination number of the thread needed can according to the CPU of computer (Central Processing Unit, Central processing unit) check figure determination, wherein, the destination number of thread is more than CPU core number, Ke Yishi Its multiple, than the thread that if any 4 cores, then can start 1.5 times of CPU core number, i.e. destination number For 6, CPU core number at least two, i.e., by multicore architecture, coding of graphics can be better achieved Parallelization, and then improve coding of graphics speed.
In any of the above-described technical scheme, it is preferable that the control module specifically includes：Detect mould Block, for detecting whether second coding unit in the previous coding unit row is completed to compile Code；Starting module, in detecting the previous coding unit row in the detection module When two coding units complete coding, start the in coding unit row described in adjacent the latter The encoding tasks of one coding unit；And judge module, for starting in the starting module The volume of other coding units in addition to first coding unit in coding unit row described in the latter Before code task, volume adjacent with other described coding units in the previous coding unit row is judged Whether code unit is completed to encode.
In the technical scheme, treated starting multiple coding unit rows successively according to default startup relation When multiple coding regions of encoded picture carry out parallel encoding, the first coding of the latter coding unit row Whether the startup of unit needs to complete to encode according to the second coding unit of adjacent previous coding unit row Depending on, i.e., the entropy code state of the first coding unit of current coded unit row is needed from adjacent previous Second coding unit of individual coding unit row obtains, if the second coding unit of previous coding unit row The first coding unit that completion coding can just start the latter coding unit row is encoded, with such Push away, to realize that parallelization encodes；Moreover, upon actuation in a coding unit row from second coding Before the encoding tasks for other coding units that unit starts, except the previous volume in this coding unit row It is outer that code unit completes coding, it is also necessary to judge in adjacent previous coding unit row with current coded unit Whether adjacent coding unit, which is completed, encodes, if so, can then start the coding of the current coded unit Task, otherwise wait for temporarily not starting coding work, to realize that parallelization encodes effectively, in an orderly manner.
In any of the above-described technical scheme, it is preferable that the control module is additionally operable to：When the target Quantity is less than the quantity of multiple coding unit rows, and each thread has correspondingly performed an institute When stating the encoding tasks of coding unit row, thread described in control release is single for performing other codings The encoding tasks of first row.
In the technical scheme, thread may be reused, and particularly be determined according to CPU core number Thread destination number be less than image encoder in coding unit row quantity when, to make full use of Existing resource completes picture parallel encoding task.
In any of the above-described technical scheme, it is preferable that also include：Processing module, at described stroke Before the picture to be encoded is divided into the multiple coding region by sub-module, compiled by the picture Code device is based on HEVC standard and carries out precoding processing to the picture to be encoded.
In the technical scheme, in order to ensure picture compression efficiency, it is necessary to treating encoded picture draw / preceding, the picture to be encoded is carried out at the precoding based on HEVC standard by image encoder Reason, to be simplified to code stream header file so that picture is more suitable for parallel encoding, in this way, The coding of graphics parallelization based on HEVC standard is realized, to ensure the same of higher compression efficiency When, coding of graphics speed is effectively improved, so as to lift Consumer's Experience.
In addition, after all threads are finished, i.e., all coding unit rows are to corresponding code area When coding is completed in domain, then each coding region after the completion of coding, shape can be integrated by processing module Into the picture after coding, terminate picture parallel encoding task.
Pass through technical scheme, it is possible to achieve the coding of graphics based on HEVC standard is parallel Change, while higher compression efficiency is ensured, to effectively improve coding of graphics speed, so as to carry Rise Consumer's Experience.
Brief description of the drawings
Fig. 1 shows the flow signal of picture parallel encoding method according to an embodiment of the invention Figure；
Fig. 2 shows first coding of different coding cell row according to an embodiment of the invention The coding of unit starts dependence schematic diagram；
Fig. 3 is shown in each coding unit row according to an embodiment of the invention except first volume The coding of other coding units outside code unit starts dependence schematic diagram；
Fig. 4 shows the block diagram of picture parallel encoder system according to an embodiment of the invention.
Embodiment
In order to be more clearly understood that the above objects, features and advantages of the present invention, with reference to attached The present invention is further described in detail for figure and embodiment.It should be noted that not In the case of conflict, the feature in embodiments herein and embodiment can be mutually combined.
Many details are elaborated in the following description to facilitate a thorough understanding of the present invention, still, The present invention can also be different from other modes described here to implement using other, therefore, the present invention Protection domain do not limited by following public specific embodiment.
Fig. 1 shows the flow signal of picture parallel encoding method according to an embodiment of the invention Figure.
As shown in figure 1, picture parallel encoding method according to an embodiment of the invention, including： Step 102, picture to be encoded is divided into multiple coding regions, each coding region is right respectively A coding unit row in image encoder is answered, and each coding unit row includes multiple volumes Code unit；Step 104, multiple coding unit rows are started to institute according to default startup relation successively State the parallel encoding of multiple coding regions.
In the technical scheme, picture to be encoded is divided into by multiple code areas by image encoder Domain, and it is right respectively by the different coding unit rows comprising multiple coding units of the image encoder Corresponding coding region is encoded, moreover, closed according to default start between multiple coding unit rows System starts coding work and carries out parallel encoding successively, so, it is possible to achieve coding of graphics parallelization, have Effect ground improves coding of graphics speed, so as to lift Consumer's Experience.
In the above-mentioned technical solutions, it is preferable that also include：Determine to be used to perform according to CPU core number The destination number of the thread of the encoding tasks of multiple coding unit rows, wherein, the destination number More than the CPU core number, and the corresponding volume for performing a coding unit row of each thread Code task.
In the technical scheme, specifically, when each coding unit row performs encoding tasks, journey is encoded For sort run on corresponding thread, i.e., each thread performs the encoding tasks of a coding unit row, and The destination number of the thread needed can according to the CPU of computer (Central Processing Unit, Central processing unit) check figure determination, wherein, the destination number of thread is more than CPU core number, Ke Yishi Its multiple, than the thread that if any 4 cores, then can start 1.5 times of CPU core number, i.e. destination number For 6, CPU core number at least two, i.e., by multicore architecture, coding of graphics can be better achieved Parallelization, and then improve coding of graphics speed.
In any of the above-described technical scheme, it is preferable that the step 104 specifically includes：Detect previous Whether second coding unit in the individual coding unit row is completed to encode；It is previous detecting When second in the individual coding unit row coding unit completes coding, start adjacent latter The encoding tasks of first coding unit in the individual coding unit row；And upon actuation one The coding of other coding units in addition to first coding unit in the individual coding unit row is appointed Before business, coding list adjacent with other described coding units in the previous coding unit row is judged Whether member is completed to encode.
In the technical scheme, treated starting multiple coding unit rows successively according to default startup relation When multiple coding regions of encoded picture carry out parallel encoding, the first coding of the latter coding unit row Whether the startup of unit needs to complete to encode according to the second coding unit of adjacent previous coding unit row Depending on, i.e., the entropy code state of the first coding unit of current coded unit row is needed from adjacent previous Second coding unit of individual coding unit row obtains, if the second coding unit of previous coding unit row The first coding unit that completion coding can just start the latter coding unit row is encoded, with such Push away, to realize that parallelization encodes, as shown in Figure 2；Moreover, upon actuation in a coding unit row Before the encoding tasks of other coding units the second coding unit, except this coding unit It is outer that previous coding unit in row completes coding, it is also necessary to judge in adjacent previous coding unit row with Whether the adjacent coding unit of current coded unit is completed to encode, if so, can then start the current volume The encoding tasks of code unit, otherwise wait for temporarily not starting coding work, as shown in figure 3, wherein, Ci,j,kOther coding units in addition to first coding unit in expression the latter coding unit row are (i.e. Current coded unit), Ci,j,k-1Represent to be located at the adjacent of same coding unit row with current coded unit Previous coding unit, Ci,j-1,k-1、Ci,j-1,k、Ci,j-1,k+1Represent and the latter coding unit row phase The coding unit adjacent with current coded unit in adjacent previous coding unit row, with effective, orderly Realize that parallelization encodes in ground.
In any of the above-described technical scheme, it is preferable that when the destination number is less than multiple codings The quantity of cell row, and each thread has correspondingly performed the coding times of a coding unit row During business, the thread is discharged, for performing the encoding tasks of other coding unit rows.
In the technical scheme, thread may be reused, and particularly be determined according to CPU core number Thread destination number be less than image encoder in coding unit row quantity when, to make full use of Existing resource completes picture parallel encoding task.
In any of the above-described technical scheme, it is preferable that described more the picture to be encoded is divided into Before individual coding region, in addition to：Treated by the image encoder based on HEVC standard to described Encoded picture carries out precoding processing.
In the technical scheme, in order to ensure picture compression efficiency, it is necessary to treating encoded picture draw / preceding, the picture to be encoded is carried out at the precoding based on HEVC standard by image encoder Reason, to be simplified to code stream header file so that picture is more suitable for parallel encoding, in this way, The coding of graphics parallelization based on HEVC standard is realized, to ensure the same of higher compression efficiency When, coding of graphics speed is effectively improved, so as to lift Consumer's Experience.
In addition, after all threads are finished, i.e., all coding unit rows are to corresponding code area When coding is completed in domain, then each coding region after the completion of integration encodes, the picture formed after coding, Terminate picture parallel encoding task.
Fig. 4 shows the block diagram of picture parallel encoder system according to an embodiment of the invention.
As shown in figure 4, picture parallel encoder system 400 according to an embodiment of the invention, bag Include：Division module 402 and control module 404.
Wherein, division module 402, for picture to be encoded to be divided into multiple coding regions, each The coding region corresponds to a coding unit row in image encoder, and each coding respectively Cell row includes multiple coding units；Control module 404, for according to default startup relation control Start the multiple volume that multiple coding unit rows obtain to the division module 402 division successively The parallel encoding in code region.
In the technical scheme, picture to be encoded is divided into by multiple code areas by image encoder Domain, and it is right respectively by the different coding unit rows comprising multiple coding units of the image encoder Corresponding coding region is encoded, moreover, closed according to default start between multiple coding unit rows System starts coding work and carries out parallel encoding successively, so, it is possible to achieve coding of graphics parallelization, have Effect ground improves coding of graphics speed, so as to lift Consumer's Experience.
In the above-mentioned technical solutions, it is preferable that also include：Determining module 406, for according to CPU Check figure determines the destination number of the thread of the encoding tasks for performing multiple coding unit rows, its In, the destination number is more than the CPU core number, and each one institute of the corresponding execution of the thread State the encoding tasks of coding unit row.
In the technical scheme, specifically, when each coding unit row performs encoding tasks, journey is encoded For sort run on corresponding thread, i.e., each thread performs the encoding tasks of a coding unit row, and The destination number of the thread needed can according to the CPU of computer (Central Processing Unit, Central processing unit) check figure determination, wherein, the destination number of thread is more than CPU core number, Ke Yishi Its multiple, than the thread that if any 4 cores, then can start 1.5 times of CPU core number, i.e. destination number For 6, CPU core number at least two, i.e., by multicore architecture, coding of graphics can be better achieved Parallelization, and then improve coding of graphics speed.
In any of the above-described technical scheme, it is preferable that the control module 404 specifically includes：Detection Module 4042, starting module 4044 and judge module 4046.
Wherein, detection module 4042, for detecting second in the previous coding unit row Whether the coding unit, which is completed, encodes；Starting module 4044, in the detection module 4042 When detecting that second coding unit in the previous coding unit row completes coding, start The encoding tasks of first coding unit in coding unit row described in adjacent the latter；And Judge module 4046, used in starting coding unit row described in the latter in the starting module 4044 Other coding units in addition to first coding unit encoding tasks before, judge previous Whether the coding unit adjacent with other described coding units is completed to encode in the coding unit row.
In the technical scheme, treated starting multiple coding unit rows successively according to default startup relation When multiple coding regions of encoded picture carry out parallel encoding, the first coding of the latter coding unit row Whether the startup of unit needs to complete to encode according to the second coding unit of adjacent previous coding unit row Depending on, i.e., the entropy code state of the first coding unit of current coded unit row is needed from adjacent previous Second coding unit of individual coding unit row obtains, if the second coding unit of previous coding unit row The first coding unit that completion coding can just start the latter coding unit row is encoded, with such Push away, to realize that parallelization encodes；Moreover, upon actuation in a coding unit row from second coding Before the encoding tasks for other coding units that unit starts, except the previous volume in this coding unit row It is outer that code unit completes coding, it is also necessary to judge in adjacent previous coding unit row with current coded unit Whether adjacent coding unit, which is completed, encodes, if so, can then start the coding of the current coded unit Task, otherwise wait for temporarily not starting coding work, to realize that parallelization encodes effectively, in an orderly manner.
In any of the above-described technical scheme, it is preferable that the control module 404 is additionally operable to：When described Destination number is less than the quantity of multiple coding unit rows, and each thread has correspondingly performed one During the encoding tasks of the individual coding unit row, thread described in control release, for performing other volumes The encoding tasks of code cell row.
In the technical scheme, thread may be reused, and particularly be determined according to CPU core number Thread destination number be less than image encoder in coding unit row quantity when, to make full use of Existing resource completes picture parallel encoding task.
In any of the above-described technical scheme, it is preferable that also include：Processing module 408, in institute State before the picture to be encoded is divided into the multiple coding region by division module 402, pass through institute State image encoder and be based on HEVC standard to the picture progress precoding processing to be encoded.
In the technical scheme, in order to ensure picture compression efficiency, it is necessary to treating encoded picture draw / preceding, the picture to be encoded is carried out at the precoding based on HEVC standard by image encoder Reason, to be simplified to code stream header file so that picture is more suitable for parallel encoding, in this way, The coding of graphics parallelization based on HEVC standard is realized, to ensure the same of higher compression efficiency When, coding of graphics speed is effectively improved, so as to lift Consumer's Experience.
In addition, after all threads are finished, i.e., all coding unit rows are to corresponding code area When coding is completed in domain, then each code area after the completion of coding can be integrated by processing module 408 Domain, the picture formed after coding, terminate picture parallel encoding task.
Technical scheme is described in detail above in association with accompanying drawing, passes through the technical side of the present invention Case, it is possible to achieve the coding of graphics parallelization based on HEVC standard, to ensure higher compression effect While rate, coding of graphics speed is effectively improved, so as to lift Consumer's Experience.
The preferred embodiments of the present invention are the foregoing is only, are not intended to limit the invention, for For those skilled in the art, the present invention can have various modifications and variations.All essences in the present invention God any modification, equivalent substitution and improvements made etc., should be included in the present invention with principle Protection domain within.

Claims (10)

  1. A kind of 1. picture parallel encoding method, it is characterised in that including：
    Picture to be encoded is divided into multiple coding regions, each coding region corresponds to picture respectively A coding unit row in encoder, and each coding unit row includes multiple coding lists Member；
    Multiple coding unit rows are started to the multiple coding region according to default startup relation successively Parallel encoding.
  2. 2. picture parallel encoding method according to claim 1, it is characterised in that also wrap Include：
    The thread of the encoding tasks for performing multiple coding unit rows is determined according to CPU core number Destination number, wherein, the destination number is more than the CPU core number, and each thread The corresponding encoding tasks for performing a coding unit row.
  3. 3. picture parallel encoding method according to claim 2, it is characterised in that described Start multiple coding unit rows successively to the parallel of the multiple coding region according to default startup relation Coding, is specifically included：
    Whether second coding unit detected in the previous coding unit row is completed to encode；
    Second coding unit in the previous coding unit row is detected completes coding When, the coding for starting first coding unit in coding unit row described in adjacent the latter is appointed Business；And
    The volume of other in addition to first coding unit in a coding unit row upon actuation Before the encoding tasks of code unit, judge single with other described codings in the previous coding unit row Whether the adjacent coding unit of member is completed to encode.
  4. 4. the picture parallel encoding method according to Claims 2 or 3, it is characterised in that
    When the destination number is less than the quantity of multiple coding unit rows, and each thread pair When should perform the encoding tasks of coding unit row, the thread is discharged, for performing The encoding tasks of other coding unit rows.
  5. 5. the picture parallel encoding method according to Claims 2 or 3, it is characterised in that Before the picture to be encoded is divided into the multiple coding region, in addition to：
    HEVC standard is based on by the image encoder to carry out at precoding the picture to be encoded Reason.
  6. A kind of 6. picture parallel encoder system, it is characterised in that including：
    Division module, for picture to be encoded to be divided into multiple coding regions, each code area Domain corresponds to a coding unit row in image encoder respectively, and is wrapped in each coding unit row Include multiple coding units；
    Control module, for starting multiple coding unit rows successively according to default startup relation control The parallel encoding for the multiple coding region for dividing to obtain to the division module.
  7. 7. picture parallel encoder system according to claim 6, it is characterised in that also wrap Include：
    Determining module, for being determined according to CPU core number for performing multiple coding unit rows The destination number of the thread of encoding tasks, wherein, the destination number is more than the CPU core number, And each corresponding encoding tasks for performing a coding unit row of the thread.
  8. 8. picture parallel encoder system according to claim 7, it is characterised in that the control Molding block specifically includes：
    Detection module, for detecting second coding unit in the previous coding unit row Whether completion encodes；
    Starting module, in detecting the previous coding unit row in the detection module When two coding units complete coding, start the in coding unit row described in adjacent the latter The encoding tasks of one coding unit；And
    Judge module, in starting coding unit row described in the latter in the starting module except the Before the encoding tasks of other coding units outside one coding unit, the previous volume is judged Whether the coding unit adjacent with other described coding units is completed to encode in code cell row.
  9. 9. the picture parallel encoder system according to claim 7 or 8, it is characterised in that institute Control module is stated to be additionally operable to：
    When the destination number is less than the quantity of multiple coding unit rows, and each thread pair When should perform the encoding tasks of coding unit row, thread described in control release, for Perform the encoding tasks of other coding unit rows.
  10. 10. the picture parallel encoder system according to claim 7 or 8, it is characterised in that also Including：
    Processing module, for the picture to be encoded to be divided into the multiple volume in the division module Before code region, HEVC standard is based on by the image encoder picture to be encoded is carried out Precoding processing.
CN201610399069.5A 2016-06-07 2016-06-07 Picture parallel encoding method and system Pending CN107483950A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610399069.5A CN107483950A (en) 2016-06-07 2016-06-07 Picture parallel encoding method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610399069.5A CN107483950A (en) 2016-06-07 2016-06-07 Picture parallel encoding method and system

Publications (1)

Publication Number Publication Date
CN107483950A true CN107483950A (en) 2017-12-15

Family

ID=60594542

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610399069.5A Pending CN107483950A (en) 2016-06-07 2016-06-07 Picture parallel encoding method and system

Country Status (1)

Country Link
CN (1) CN107483950A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108347613A (en) * 2017-01-25 2018-07-31 龙芯中科技术有限公司 Image macro parallel encoding method and apparatus
CN110536140A (en) * 2018-09-20 2019-12-03 北京达佳互联信息技术有限公司 Method for video coding and device
CN110958443A (en) * 2019-12-16 2020-04-03 宁波大学 Fast encoding method for 360-degree video interframes

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103974081A (en) * 2014-05-08 2014-08-06 杭州同尊信息技术有限公司 HEVC coding method based on multi-core processor Tilera
CN104081781A (en) * 2012-01-20 2014-10-01 弗兰霍菲尔运输应用研究公司 Coding concept allowing parallel processing, transport demultiplexer and video bitstream
CN104780377A (en) * 2015-03-18 2015-07-15 同济大学 Parallel high efficiency video coding (HEVC) system and method based on distributed computer system
WO2015194394A1 (en) * 2014-06-20 2015-12-23 ソニー株式会社 Image encoding apparatus and method, and image decoding apparatus and method
CN105230022A (en) * 2013-05-31 2016-01-06 高通股份有限公司 Use based on the disparity vector of neighbor derive for 3D video coding and derivation disparity vector of passing through
CN105245897A (en) * 2014-03-20 2016-01-13 南京云岩信息科技有限公司 Efficient software for transcoding to hevc on multi-core processors

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104081781A (en) * 2012-01-20 2014-10-01 弗兰霍菲尔运输应用研究公司 Coding concept allowing parallel processing, transport demultiplexer and video bitstream
CN105230022A (en) * 2013-05-31 2016-01-06 高通股份有限公司 Use based on the disparity vector of neighbor derive for 3D video coding and derivation disparity vector of passing through
CN105245897A (en) * 2014-03-20 2016-01-13 南京云岩信息科技有限公司 Efficient software for transcoding to hevc on multi-core processors
CN103974081A (en) * 2014-05-08 2014-08-06 杭州同尊信息技术有限公司 HEVC coding method based on multi-core processor Tilera
WO2015194394A1 (en) * 2014-06-20 2015-12-23 ソニー株式会社 Image encoding apparatus and method, and image decoding apparatus and method
CN104780377A (en) * 2015-03-18 2015-07-15 同济大学 Parallel high efficiency video coding (HEVC) system and method based on distributed computer system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
FÉLIX HENRY: "Wavefront Parallel Processing", 《JOINT COLLABORATIVE TEAM ON VIDEO CODING (JCT-VC)OF ITU-T SG16 WP3 AND ISO/IEC JTC1/SC29/WG11 》 *
杨杭军: "基于多核处理器的视频编解码并行算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108347613A (en) * 2017-01-25 2018-07-31 龙芯中科技术有限公司 Image macro parallel encoding method and apparatus
CN108347613B (en) * 2017-01-25 2020-05-26 龙芯中科技术有限公司 Image macro block parallel coding method and device
CN110536140A (en) * 2018-09-20 2019-12-03 北京达佳互联信息技术有限公司 Method for video coding and device
CN110536140B (en) * 2018-09-20 2021-09-07 北京达佳互联信息技术有限公司 Video coding method and device
CN110958443A (en) * 2019-12-16 2020-04-03 宁波大学 Fast encoding method for 360-degree video interframes
CN110958443B (en) * 2019-12-16 2021-06-29 宁波大学 Fast encoding method for 360-degree video interframes

Similar Documents

Publication Publication Date Title
RU2018104096A (en) METHOD AND DEVICE FOR CODING OR DECODING PIXEL BLOCKS
CN103262534B (en) Improve entropy coding and decoded method and device
CN110689599B (en) 3D visual saliency prediction method based on non-local enhancement generation countermeasure network
CN107483950A (en) Picture parallel encoding method and system
CN103918270B (en) System, method and computer program product for Video coding pipeline
CN107071457A (en) Video encoding/decoding method
CN108337511A (en) Image encoding apparatus
CN109213594A (en) Method, apparatus, equipment and the computer storage medium that resource is seized
CN104539949B (en) The method and device of quick partitioning based on edge direction in HEVC screen codings
CN109788285A (en) The context model choosing method and device of a kind of quantization parameter end mark position
CN110383838A (en) Transformation limitation based on block size
CN105554591A (en) Video analysis method and device
CN103297768B (en) For video and the parallel pyramid entropy code of compression of images
KR101673186B1 (en) Apparatus and method of processing in parallel of encoding and decoding of image data by using correlation of macroblock
CN111553471A (en) Data analysis processing method and device
US20190279330A1 (en) Watermark embedding method and apparatus
Zhang et al. Implementation and improvement of wavefront parallel processing for HEVC encoding on many-core platform
US20170054990A1 (en) Encoding device, decoding device, and image processing device
KR20200119372A (en) Artificial Neural Network Based Object Region Detection Method, Device and Computer Program Thereof
CN105122806B (en) Information processing unit, control method and storage medium
CN104618715B (en) A kind of method and device obtaining minimum rate distortion costs
CN103763569B (en) A kind of HEVC fine grained parallel Forecasting Methodology based on First Input First Output
CN115661276A (en) Image data encoding method, device, apparatus, medium, and program
KR20170002512A (en) Techniques for processing a video frame by a video encoder
CN112437308A (en) WebP coding method and device

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20171215