CN107404653A - A kind of Parking quick determination method of HEVC code streams - Google Patents
A kind of Parking quick determination method of HEVC code streams Download PDFInfo
- Publication number
- CN107404653A CN107404653A CN201710367626.XA CN201710367626A CN107404653A CN 107404653 A CN107404653 A CN 107404653A CN 201710367626 A CN201710367626 A CN 201710367626A CN 107404653 A CN107404653 A CN 107404653A
- Authority
- CN
- China
- Prior art keywords
- mrow
- motion vector
- mtd
- coding unit
- flag
- 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.)
- Granted
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
- H04N19/44—Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder
-
- 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
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/119—Adaptive subdivision aspects, e.g. subdivision of a picture into rectangular or non-rectangular coding blocks
-
- 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
- H04N19/136—Incoming video signal characteristics or properties
-
- 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
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
- H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
- H04N19/51—Motion estimation or motion compensation
- H04N19/513—Processing of motion vectors
-
- 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
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
- H04N19/593—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving spatial prediction techniques
-
- 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
- H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
- H04N19/96—Tree coding, e.g. quad-tree coding
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses a kind of Parking quick determination method based on HEVC code streams, this method extracts the information such as image coding unit partition mode, motion vector, intra prediction by partial decoding of h compressed video bit stream.On this basis, there is provided a kind of intraframe coding unit virtual motion vector distribution method based on intraframe coding dividing elements and prediction residual.This method can effectively make up the intraframe coding unit of the without motion vector included in motion vector field, and good place mat has been made for the processing of subsequent motion vector field.Then the motion vector field relatively stablized by pretreatment operations such as motion vector field spatial criterion, temporal criterions.The quick detection of Parking is finally carried out according to the relation of moving target motion vector in adjacent two frames video.Method proposed by the present invention can fast and effectively detect the Parking in traffic surveillance videos under conditions of endless full decoder HEVC video compression datas.
Description
Technical field
The invention belongs to the field of Video processing, is related to the vehicle behavior event contained in the compressed domain to video bag and knows
Not, and in particular to a kind of Parking quick determination method of HEVC code streams.
Background technology
The automatic discrimination of monitoring and the behavior of vehicle has important meaning in terms of prevention traffic accident, guarantee property safety
Justice.Currently mainly monitoring camera is set to be monitored road vehicle by each traffic intersection, advanced traffic monitoring regards
Frequency has the characteristics that definition is high, data volume is big.Vehicle behavioral value technology be exactly using computer how automatic identification video
In the vehicle behavior event that includes.
The behavior of vehicle is generally divided into individual vehicle behavior and the accumulation behavior of a large amount of vehicles.In terms of the Activity recognition of vehicle
Research is concentrated mainly in pixel domain, and the video code flow after will encoding first is decoded to pixel domain, then passes through the place to pixel
Manage further recognition detection vehicle behavior.The vehicle behavioral value of pixel domain is broadly divided into two classes:Activity recognition based on track
Method and the Activity recognition method based on topic model.The former main research object is the behavior of individual vehicle, and the latter is main
Research object be the accumulation of a large amount of vehicles behavior.
Vehicle behavioral value based on video belongs to an application of Video Events processing, although at current major video
Reason technology is pixel domain processing method, but also result in mitigate the compressed domain video treatment technology of computation complexity
The concern of person.What it is due to com- pressed-domain processing technical finesse is data after video compress, and its data volume is relatively much smaller, simultaneously
Effectively usable information content is also relatively fewer, therefore certainly exists the problem of event detection precision is low.
Vehicle behavior event detection technology in the video of conventional pixel domain is broadly divided into three modules, i.e. moving vehicle target
Detection, target vehicle tracking, track fitting and behavior event analysis.It is actually to examine the vehicle behavior event in video
Survey is divided into the standalone modules such as object detection and recognition traditional in computer vision technique, motion target tracking and trajectory analysis
Studied and realized respectively.And the Video processing research in current compression domain is concentrated mainly on moving Object Segmentation, target identification
And tracking etc., therefore the vehicle behavior event detection of compression domain can also continue to use the vehicle behavior event inspection of pixel domain
Examining system, but the vehicle characteristics and environmental characteristic information that can be provided in compression domain are fewer, cause to be owned in pixel domain
Vehicle behavior event detecting method be all effective.But compression domain include some pixel domains without useful information, such as coding singly
The division of member, the motion vector of predictive coding unit, prediction residual etc..HEVC video encoding standards are newest video compress
Standard, future will be widely used.Therefore the compression information such as division of the coding unit by directly being provided using HEVC code streams
Realize the detection of vehicle parking event, it becomes possible to avoid above mentioned problem.But there is presently no disclose this detection in the prior art
Method.
The content of the invention
Present invention aims at directly utilize motion vector information in HEVC code streams, coding unit division, intraframe coding list
The information such as first prediction residual carry out vehicle parking event detection.A kind of HEVC code streams are provided and carry out Parking quick detection side
Method, unnecessary video decoding process is avoided, so as to reduce the complexity of calculating, and do not destroying original compression video counts
Quick vehicle parking event detection is carried out on the basis of.
The technical scheme that uses to achieve these goals of the present invention be to the compressed video stream data that will analyze (such as
HEVC bit stream datas) partial decoding of h analysis is carried out, handled by simple entropy decoding, extract the fortune included in bit stream data
The information such as dynamic vector, coding unit division, residual error data, confluence analysis is carried out to these data, extracts the feature of vehicular events
Information, then vehicle parking behavior event is detected by characteristic.It is of the invention mainly to utilize in parking behavior event
During generation, the motion vector in vehicle target region is what this entirely different feature was carried out in adjacent video frames, that is, thing of stopping
Before part occurs, the motion vector in vehicle target region is usually bigger.After parking, the motion vector in vehicle target region
Generally zero motion vector.
Above-mentioned technical proposal specifically, comprises the following steps:
Step 1:Coding unit structure, motion vector and the intraframe coding of kth frame encoded video are extracted from HEVC code streams
Unit residual error data;
Step 2:It is empty for the configuration of each intraframe coding unit according to coding unit structure and intraframe coding unit residual error data
Intend motion vector;
Step 3:Different size of coding unit is uniformly divided into NxN square, and by the fortune of coding unit before segmentation
Dynamic vector is that each calibrated bolck configures motion vector, and the motion vector of different reference frames then is converted into unified reference frame
Motion vector, form motion vector field
Step 4:To motion vector fieldDenoising is carried out, is formed new
Step 5:Given threshold Thstart, for distinguishing motion vector fieldThe background block and moving vehicle block of Plays block,
Method is as follows:Mark a flag_start, flag_start, which are configured, for each calibrated bolck represents that the calibrated bolck belongs to for 1
Moving vehicle block, flag_start represent that the calibrated bolck belongs to background block for 0, judge motion vector fieldIn each calibrated bolck
Bi k, i expressionsThe sequence number of Plays block, k expression frame numbers, motion vector MV, if | MV | >=Thstart, then the calibrated bolck
Flag flag_start is set to 1, and otherwise, flag_start is set to 0, by motion vector fieldIn each flag_start
It is set to all pixels of the calibrated bolck that 1, flag_start is 0 for all pixels of 1 calibrated bolck and is set to 0, and counts flag_
Start is 1 calibrated bolck number N umstart;
Step 6:The coding unit structure of extraction k+1 frame video datas, motion vector, intraframe coding list from HEVC code streams
First residual error data, motion vector processing is carried out according to above-mentioned steps 2 to step 4, obtains motion vector fieldWillMiddle motion arrow
Amount MV amplitude is more than or equal to threshold value ThstartCalibrated bolck Bi k+1Kth frame, which is moved to, by MV retrospects obtains Bm k, and willMiddle Bm kIt is interior
Bi k+1The pixel of block correspondence position is set to 0, and other positions pixel is constant;
Step 7:Threshold value Th is setstop, for distinguishing the type of calibrated bolck, and a mark is configured for each calibrated bolck
Flag_end, flag_end represent that the block belongs to possible parking area for 1, and flag_end represents that the block belongs to non-parking area for 0
Domain, the calibrated bolck that all flag_start of traversal kth frame are 1It is 1 to count flag_startMiddle pixel value is 1 picture
Vegetarian refreshments number N umnext, calculate NumnextWithSize block_size percentage, determine the type of calibrated bolck, calculating side
Method is as follows:
Step 8:AskMiddle flag_end is 1 calibrated bolck number N umstopIt is 1 region minimum enclosed rectangle with flag_end
Include calibrated bolck number N umtotalRatio, and realize according to following formula the judgement of Parking:
State_car represents that this has Parking for 1, and state_car is that Parking, state_ is not present in 0 expression
Car is 1, and detection terminates, state_car 0, removes a frame data, continues step 1 and arrives step 8.
Further, described in above-mentioned steps 2 of the present invention it is each intraframe coding unit configuration virtual motion vector, including with
Lower step:
Step 1:Initialization frame interior coding unit failure allocation result counts mark Intra_flag=0, and according in frame
Coding unit division size is divided into following grade:
The intraframe coding unit that note does not configure motion vector is BnoMV;
Step 2:Travel through BnoMV, and the prediction residual res of intraframe coding unit is calculated, specific computational methods are as follows:
(1) the residual error data r of pixel in current intraframe coding unit is readi k, wherein k expression current coded units are whole
The index ID of individual division unit, it is integer, i represent current pixel point current intraframe coding unit index ID, i it is big
Small is 1 to arrive current coded unit size c_size;
(2) residual error of current coded unit is calculatedWherein
Step 3:Virtual motion vector is distributed to intraframe coding unit, specific distribution method is as follows:
(1) if BnoMVIntra_level be less than or equal to ThB, then by BnoMVIt is considered as background environment part, while will be right
Answer BnoMVMotion vector be set to (0,0);
(2) if BnoMVIntra_level be more than or equal to ThFOr BnoMVResidual error res be more than threshold value Thres, then will
BnoMVIt is considered as moving vehicle part, meanwhile, extract BnoMVNon-zero motion vectors are made by the motion vector of surrounding interframe encode unit
Median motion vector is extracted with polar coordinates median filtering method, current B is assigned with thisnoMVBlock, if adjacent encoder unit is all
Motion vector using intra-frame encoding mode or neighboring prediction coding unit is zero, then retains current intraframe coding unit mould
Formula, virtual motion vector wouldn't be configured;
(3) if BnoMVIntra_level be more than ThBAnd it is less than ThF, while res is less than Thres, then B is extractednoMVWeek
The motion vector of adjacent interframe coding unit is enclosed, extracting median motion using polar coordinates median filtering method to motion vector swears
Amount, assigns current BnoMVBlock, if adjacent encoder unit is all intra-frame encoding mode, retain current intraframe coding unit mould
Formula, virtual motion vector wouldn't be configured;
Step 4:Judge whether the distribution of intraframe coding unit motion vector succeeds, Intra_ is made if distribution is failed
Flag=Intra_flag+1;Otherwise, next step is performed;
Step 5:Judge whether current intraframe coding unit travels through completion, if current intraframe coding unit has not traveled through
Into then jump procedure 2;Otherwise, next step is performed;
Step 6:If Intra_flag > 0, jump procedure 1;Otherwise circulation is terminated.
Further, use polar coordinates median filtering method of the present invention extracts median motion vector and includes following step
Suddenly:
Step 1:By the inter prediction encoding unit universal formulation in the current neighborhood of intraframe coding unit 4 to 4 × 4 sizes
Cell level, and the motion vector after division is allocated according to the motion vector of original predictive unit;
Step 2:Motion vector after division is saved in internal memory, asks for the angle, θ of all motion vectorsi, range value |
|vi||2, and respectively by both by sequence from small to large;
Step 3:Angle, θ is asked for using medium filteringiAnd range value | | vi||2Intermediate value, and using the median vector as work as
The motion vector of preceding intraframe coding unit.
Further, motion vector is handled described in above-mentioned steps 3 of the present invention, composition motion vector field, including with
Lower step:
Step 1:The processing of motion vector Spatial normalization, different size coding unit and corresponding motion vector are divided into
The calibrated bolck and respective motion vectors of unified size 4 × 4, method are as follows:It is more than 4 × 4 part for coding unit size,
The coding unit is divided into the calibrated bolck of 4 × 4 sizes, the motion vector of all calibrated bolcks is the coding unit fortune before division
Dynamic vector;
Step 2:The processing of motion vector temporal criterionization, the motion vector based on multi-reference frame extracted in HEVC is converted
The motion vector of calibrated bolck is corresponded to for same adjacent video frames, it is assumed that current video P frames are designated as Ic, IcIn i-th prediction compile
Code unit is designated as Bi, predictive coding unit BiMotion vector be designated as MV (Bi), its reference frame is designated as Ir, subscript c, i, r here
Identified for integer number, the motion vector of temporal criterion
Further, the motion vector field denoising described in above-mentioned steps 4 of the present invention, comprises the following steps:
Step 1:Calculate motion vector MV horizontal MVxWith longitudinal component MVyAbsolute value sum SumMV, according to SumMVWith
Threshold value T comparison decision MV value, method are as follows:
T selection is relevant with the motion estimation techniques used in HEVC video encoding standards;
Step 2:Using 3 × 3 templates pairMotion vector carry out median filter process.
Further, the region minimum enclosed rectangle described in calculation procedure 8 of the present invention includes calibrated bolck number N umtotal's
Method is:
Step 1:Ask for motion vector fieldMiddle flag_end flags are 1 calibrated bolck number N umstop;
Step 2:Ask for including the minimum enclosed rectangle rect for the calibrated bolck that flag_end flags are 1;
Step 3:Ask for the number N um for all calibrated bolcks that rect is includedtotal。
Compared with prior art, beneficial outcomes of the invention:
1, the present invention does not need complete decoding compressed video data, when being significantly reduced a large amount of calculating that video decoding needs
Between;Processing data is mainly motion vector simultaneously, greatly reduces the complexity issue of calculating.
2, the motion vector distribution of the invention according to moving vehicle in video, it is not necessary to carry out vehicle target detection and tracking
Deng the calculation procedure of complexity, the detection of Parking is directly realized by, there is good application prospect.
Brief description of the drawings
Vehicle parking event quick detection block diagram in Fig. 1 HEVC code streams.
The partition structure schematic diagram of Fig. 2 HEVC code tree units.
The data structure result figure extracted in Fig. 3 super expressway monitor video code streams.
Intraframe coding unit motion vectors of the Fig. 4 based on coding unit division and prediction residual distributes block diagram.
Fig. 5 intraframe codings modular movement vector distributes schematic diagram.
Fig. 6 motion vectors and coding unit division schematic diagram.
Fig. 7 motion vectors temporal criterionization handles schematic diagram.
Fig. 8 moves block matching schematic diagram.
Fig. 9 minimum enclosed rectangle schematic diagrames.
Embodiment
The invention provides a kind of quick Parking detection algorithm based on HEVC code streams, HEVC code stream numbers are not being decoded
The quick detection of Parking is realized on the premise of.Meanwhile one kind is provided in data processing and is based on coding unit
Division and intraframe coding unit prediction residual are the method that intraframe coding unit configures virtual motion vector, realize Parking
Quick detection.
As shown in figure 1, specific implementation needs to perform following steps:
Step 101, extract from HEVC code streams and compiled in coding unit structure, motion vector and the frame of kth frame encoded video
Code unit residual error data.The code stream position according to corresponding to HEVC video compression coding standard syntax semantemes orient information above,
Then the information is decoded by steps such as entropy decodings.Fig. 2 is that the code tree dividing elements structure of HEVC video encoding standards is shown
It is intended to, HEVC uses coding unit tissue structure of the quad-tree structure as frame of video, and minimum coding unit (CU) size is
8×8.Coding unit is as shown in Figure 2.The CU that wherein size is 16 × 16 in fig. 2 can continue to be divided according to division rule
For 8 × 8.And the division of predicting unit (PU) and converter unit (TU) is further on the basis of CU in HEVC coding standards
Division, but its minimum division unit size is 4 × 4.Fig. 3 is that experiment extracts drawing in certain high speed monitor video code stream
Point result figure, grey arrow represents the size and Orientation of motion vector in figure, and black box be coding unit division result, is tested
In due to HEVC inter-frame prediction techniques employ a quarter pixel precision processing, i.e., by interpolation processing operation to adjacent picture
Pixel filling is carried out between element, so the motion vector information of extraction is about four times of original image pixels size;
Step 102, according to coding unit structure and intraframe coding unit residual error data, configured for each intraframe coding unit
Virtual motion vector.As shown in figure 4, concrete configuration process is as follows:
Step 401, initialization frame interior coding unit failure allocation result counts mark Intra_flag=0.And according to frame
Interior coding dividing elements size is divided into following grade:
Here P () represents the division size of intraframe coding unit, and 1-5 represents to be compiled according to intraframe coding dividing elements respectively
Code cell level.After by intraframe coding dividing elements for five grades, the resolution ratio of video is read.Regarded for different
Frequency division resolution and monitor video, different threshold value Th is setBAnd ThF.According to the comparison of the grade of intraframe coding unit and threshold value,
Preliminary division target code unit belongs to moving vehicle prospect or background environment.Note is not configured in the frame of motion vector and compiled
Code unit is BnoMV;
Step 402, B is traveled throughnoMV, and calculate the prediction residual res of intraframe coding unit.Specific computational methods are as follows:
(1) the residual error data r of pixel in current intraframe coding unit is readi k, wherein k expression current coded units are whole
The index ID (for integer) of individual division unit, i represent index ID (i size of the current pixel point in current intraframe coding unit
Current coded unit size c_size is arrived for 1);
(2) residual error of current coded unit is calculatedWherein
Step 403, virtual motion vector is distributed to intraframe coding unit.Specific distribution method is as follows:
(1) if BnoMVIntra_level be less than or equal to ThB, then by BnoMVIt is considered as background environment part, while will be right
Answer BnoMVMotion vector be set to (0,0);
(2) if BnoMVIntra_level be more than or equal to ThFOr BnoMVResidual error res be more than threshold value Thres, then will
BnoMVIt is considered as moving vehicle part.Meanwhile extract BnoMVNon-zero motion vectors are made by the motion vector of surrounding interframe encode unit
Median motion vector is extracted with polar coordinates median filtering method, current B is assigned with thisnoMVBlock.If adjacent encoder unit is all
Motion vector using intra-frame encoding mode or neighboring prediction coding unit is zero, then retains current intraframe coding unit mould
Formula, virtual motion vector wouldn't be configured.Residual error is more than the number of pixels of certain threshold value in wherein res expression current coded units,
ThresRepresent current coded unit whether be target vehicle square threshold value, because the size of coding unit is to differ in practice
Cause, so threshold value Th hereresAlso different settings should be carried out according to different coding unit sizes, here by ThresTurn
Turn to Thres=block_size × th, th ∈ (0,1), it is 0.3 that th is chosen in experiment;
(3) if BnoMVIntra_level be more than ThBAnd it is less than ThF, while res is less than Thres, then B is extractednoMVWeek
The motion vector of adjacent interframe coding unit is enclosed, extracting median motion using polar coordinates median filtering method to motion vector swears
Amount, assigns current BnoMVBlock.If adjacent encoder unit is all intra-frame encoding mode, retain current intraframe coding unit mould
Formula, virtual motion vector wouldn't be configured.
Step 404, judge whether the distribution of intraframe coding unit motion vector succeeds.Made if distribution is failed
Intra_flag=Intra_flag+1.Otherwise, next step is performed;
Step 405, judge whether current intraframe coding unit travels through completion.If current intraframe coding unit has not traveled through
Into then jump procedure 2.Otherwise, next step is performed;
Step 406, if Intra_flag > 0, jump procedure 1;Otherwise circulation is terminated.
It is as shown in Figure 5 for the polar coordinates median filter method used in step 403, specific calculating process.It is specific to calculate
Process is as follows:
MV storage list list is expressed as V=(v in Fig. 51,v2,v3,v3,v4,v4,v5,v5,v6,v6,v6,v6).Wherein
v6Occur 4 times, because its coding unit can be divided into 4 minimum code squares adjacent with pending intraframe coding unit
(such as 4 × 4 squares chosen in testing).Similarly, v3,v4,v5Occur twice respectively, and v1,v2Only occur once.This MV
List can at most include 16 MV, i.e., pending intraframe coding unit it is adjacent be interframe encode unit entirely.And treated in Fig. 5
The B of processingnoMVIntraframe coding unit be present in adjacent encoder unit.
Assuming that V=(vi)I=1:nIt is input motion vector, if θiRepresent motion vector viAnd vi+1Between angle (it is assumed here that
v1≡vn+1).Then it can define and m motion vector composition set is chosen from motion vector setWherein
viChosen from input motion set of vectors, and m and I can be expressed as beWithFor the meaning rounded downwards.New motion vector set is understood from above-mentioned two formulaComprising
Angle and minimum half motion vector in original motion vector set.It is finally the MV of pending intraframe coding unit distribution
It can be asked for by following two mathematic(al) representationWith
Step 103, different size of coding unit is uniformly divided into 4x4 square (hereinafter referred to as calibrated bolck, calibrated bolck
Size block_size can be 4x4 or 8x8, specifically be determined by the division of HEVC coding units, it is of the invention using 4x4 sizes as
Example), and motion vector is configured for each calibrated bolck by the motion vector of coding unit before segmentation, then by the fortune of different reference frames
Dynamic vector is converted to the motion vector (present invention uses neighboring reference frame) of unified reference frame, forms motion vector fieldSpecifically
Calculating process is broadly divided into motion vector Spatial normalization and time standard:
(1) motion vector Spatial normalization is handled.As shown in fig. 6, black bars are intraframe coding unit in figure, so its
In the absence of motion vector.And do not have in figure arrow coding unit be not represent the coding unit motion vector is not present, and
It is to say that the motion vector of the coding unit is zero motion vector.The coding unit of all sizes and corresponding motion vector are turned
Turn to the calibrated bolck and respective motion vectors of unified size 4 × 4.Specifically, it is more than 4 × 4 portion for coding unit size
Point, the coding unit is converted into the calibrated bolck of 4 × 4 sizes, the motion vector of all substandard blocks is the coding list before division
First motion vector.As shown in fig. 6, the processing more than, first will be to intraframe coding unit assigned motion vector, then by Fig. 6
(a) the different coding unit shown in is converted into the calibrated bolck of unified size shown in Fig. 6 (b), and motion vector corresponding to distribution.
The calibrated bolck for not having motion vector arrow to identify in figure is because its motion vector is zero motion vector;
(2) motion vector temporal criterionization is handled.As shown in fig. 7, HEVC inter frame motion estimations are handled using multi-reference frame
Technology.Current video P frames are designated as Ic, IcIn the i-th square be designated as Bi, square BiMotion vector be designated as MV (Bi), it is joined
Examine frame and be designated as Ir, subscript c, i, r identify for integer number here.The motion vector of temporal criterionization processing
Performance is in fig. 7, it is the motion vector of different reference frames to be converted to the motion vector of same reference frame.
Step 104, to motion vector fieldDenoising is carried out, is formed newSpecific calculating process:
(1) motion vector MV horizontal and vertical component (MV is calculatedxAnd MVy) absolute value and SumMV, according to SumMVWith
Threshold value T comparison determines MV value, specific as follows:
T selection is relevant with the motion estimation techniques used in HEVC video encoding standards, such as a quarter pixel motion
Estimation, the estimation of half pixel motion etc. will cause T difference;
(2) 3 × 3 size templates pair are usedCarry out median filter process.Centered on current square, in 3 × 3 squares
Motion vector sorted from small to large with MV range values, from median replace current square motion vector.
Step 105, given threshold Thstart, for distinguishing motion vector fieldThe background block and moving vehicle of Plays block
Block.Method is as follows:Mark a flag_start, flag_start, which are configured, for each calibrated bolck represents the calibrated bolck category for 1
In moving vehicle block, flag_start represents that the calibrated bolck belongs to background block for 0.Judge motion vector fieldIn each standard
Block Bi k(i is representedThe sequence number of Plays block, k represent frame number) motion vector MV, if MV >=Thstart, then the calibrated bolck
Flag flag_start is set to 1, and otherwise, flag_start is set to 0.By dynamic vector fieldIn each flag_start be 1
The all pixels of calibrated bolck be set to all pixels of calibrated bolck that 1, flag_start is 0 and be set to 0.And count flag_
Start is 1 calibrated bolck number N umstart;
Step 106, the coding unit structure of extraction k+1 frame video datas, motion vector, intraframe coding from HEVC code streams
Unit residual error data.Motion vector processing is carried out according to above-mentioned steps 2 to step 4, obtains motion vector fieldWillMiddle motion
Vector MV amplitude is more than or equal to threshold value ThstartCalibrated bolck Bi k+1Kth frame, which is moved to, by MV retrospects obtains Bm k, and willIn
Bm kInterior Bi k+1The pixel of block correspondence position is set to 0, and other positions pixel is constant.It is specific as shown in figure 8, black mark in k+1 frames
The quasi- pending motion square of block, is moved to k frames by motion vector MV by it and obtains black standard block, the block and B1, B2, B3, B4
There is common factor, intersecting partial pixel position is set to 0, other positions pixel is constant;;
Step 107, threshold value Th is setstop, for distinguishing the type of calibrated bolck, and a mark is configured for each calibrated bolck
Flag_end, flag_end represent that the block belongs to possible parking area for 1, and flag_end represents that the block belongs to non-parking area for 0
Domain.Travel through the calibrated bolck that all flag_start of kth frame are 1It is 1 to count flag_startMiddle pixel value is 1 picture
Vegetarian refreshments number N umnext, calculate NumnextWithSize block_size percentage, determine the type of calibrated bolck, calculating side
Method is as follows:
Step 108, as shown in figure 9, asking forMiddle flag_end is 1 calibrated bolck number N umstopIt is 1 with flag_end
Region minimum enclosed rectangle includes calibrated bolck number N umtotalPercentage, in fig.9, the calibrated bolck of gray area part is
Flag_end is 1 calibrated bolck, so Num in the figurestopFor 12.The calibrated bolck that the region minimum enclosed rectangle includes then wraps
White and grey two parts calibrated bolck are included, Num in figuretotalFor 25.And the judgement of Parking is realized according to following formula:
State_car represents that this has Parking for 1, and state_car is that Parking is not present in 0 expression.The present invention
Middle selected threshold ThcarFor 0.5, ThsFor 0.65.Here state_car is that the 1 expression square is Parking, state_car
Represent Parking not be present for 0.
Claims (6)
1. a kind of Parking quick determination method based on HEVC code streams, it is characterised in that comprise the following steps:
Step 1:Coding unit structure, motion vector and the intraframe coding unit of kth frame encoded video are extracted from HEVC code streams
Residual error data;
Step 2:According to coding unit structure and intraframe coding unit residual error data, for the virtual fortune of each intraframe coding unit configuration
Dynamic vector;
Step 3:Different size of coding unit is uniformly divided into NxN square, and by the motion arrow of coding unit before segmentation
Measure and configure motion vector for each calibrated bolck, then motion that the motion vector of different reference frames is converted to unified reference frame is sweared
Amount, form motion vector field
Step 4:To motion vector fieldDenoising is carried out, is formed new
Step 5:Given threshold Thstart, for distinguishing motion vector fieldThe background block and moving vehicle block of Plays block, method
It is as follows:Mark a flag_start, flag_start, which are configured, for each calibrated bolck represents that the calibrated bolck belongs to sport(s) car for 1
Block, flag_start are that the 0 expression calibrated bolck belongs to background block, judge motion vector fieldIn each calibrated bolck Bi k, i expressionsThe sequence number of Plays block, k expression frame numbers, motion vector MV, if | MV | >=Thstart, then the flag flag_ of the calibrated bolck
Start is set to 1, and otherwise, flag_start is set to 0, by motion vector fieldIn each flag_start be 1 calibrated bolck
The all pixels that all pixels are set to the calibrated bolck that 1, flag_start is 0 are set to 0, and count the standard that flag_start is 1
Block number mesh Numstart;
Step 6:The coding unit structure of extraction k+1 frame video datas, motion vector, intraframe coding unit are residual from HEVC code streams
Difference data, motion vector processing is carried out according to above-mentioned steps 2 to step 4, obtains motion vector fieldWillMiddle motion vector MV
Amplitude be more than or equal to threshold value ThstartCalibrated bolck Bi k+1Kth frame, which is moved to, by MV retrospects obtains Bm k, and willMiddle Bm kInterior Bi k+1
The pixel of block correspondence position is set to 0, and other positions pixel is constant;
Step 7:Threshold value Th is setstop, for distinguishing the type of calibrated bolck, and a mark flag_ is configured for each calibrated bolck
End, flag_end represent that the block belongs to possible parking area for 1, and flag_end represents that the block belongs to non-parking area for 0, time
Go through the calibrated bolck that all flag_start of kth frame are 1It is 1 to count flag_startMiddle pixel value is 1 pixel number
Mesh Numnext, calculate NumnextWithSize block_size percentage, determines the type of calibrated bolck, and computational methods are as follows:
<mrow>
<mi>f</mi>
<mi>l</mi>
<mi>a</mi>
<mi>g</mi>
<mo>_</mo>
<mi>e</mi>
<mi>n</mi>
<mi>d</mi>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
<mtd>
<mrow>
<mfrac>
<mrow>
<msub>
<mi>Num</mi>
<mrow>
<mi>n</mi>
<mi>e</mi>
<mi>x</mi>
<mi>t</mi>
</mrow>
</msub>
</mrow>
<mrow>
<mi>b</mi>
<mi>l</mi>
<mi>o</mi>
<mi>c</mi>
<mi>k</mi>
<mo>_</mo>
<mi>s</mi>
<mi>i</mi>
<mi>z</mi>
<mi>e</mi>
</mrow>
</mfrac>
<mo>&GreaterEqual;</mo>
<msub>
<mi>Th</mi>
<mrow>
<mi>s</mi>
<mi>t</mi>
<mi>o</mi>
<mi>p</mi>
</mrow>
</msub>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
<mtd>
<mrow>
<mi>o</mi>
<mi>t</mi>
<mi>h</mi>
<mi>e</mi>
<mi>r</mi>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
Step 8:AskMiddle flag_end is 1 calibrated bolck number N umstopIt is that 1 region minimum enclosed rectangle includes with flag_end
Calibrated bolck number N umtotalRatio, and realize according to following formula the judgement of Parking:
<mrow>
<mi>s</mi>
<mi>t</mi>
<mi>a</mi>
<mi>t</mi>
<mi>e</mi>
<mo>_</mo>
<mi>c</mi>
<mi>a</mi>
<mi>r</mi>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
<mtd>
<mtable>
<mtr>
<mtd>
<mrow>
<mfrac>
<mrow>
<msub>
<mi>Num</mi>
<mrow>
<mi>s</mi>
<mi>t</mi>
<mi>o</mi>
<mi>p</mi>
</mrow>
</msub>
</mrow>
<mrow>
<msub>
<mi>Num</mi>
<mrow>
<mi>t</mi>
<mi>o</mi>
<mi>t</mi>
<mi>a</mi>
<mi>l</mi>
</mrow>
</msub>
</mrow>
</mfrac>
<mo>&GreaterEqual;</mo>
<msub>
<mi>Th</mi>
<mrow>
<mi>c</mi>
<mi>a</mi>
<mi>r</mi>
</mrow>
</msub>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>a</mi>
<mi>n</mi>
<mi>d</mi>
</mrow>
</mtd>
<mtd>
<mrow>
<mfrac>
<mrow>
<msub>
<mi>Num</mi>
<mrow>
<mi>s</mi>
<mi>t</mi>
<mi>o</mi>
<mi>p</mi>
</mrow>
</msub>
</mrow>
<mrow>
<msub>
<mi>Num</mi>
<mrow>
<mi>s</mi>
<mi>t</mi>
<mi>a</mi>
<mi>r</mi>
<mi>t</mi>
</mrow>
</msub>
</mrow>
</mfrac>
<mo>&GreaterEqual;</mo>
<msub>
<mi>Th</mi>
<mi>s</mi>
</msub>
</mrow>
</mtd>
</mtr>
</mtable>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
<mtd>
<mrow>
<mi>o</mi>
<mi>t</mi>
<mi>h</mi>
<mi>e</mi>
<mi>r</mi>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
State_car represents that this has Parking for 1, and state_car is that Parking is not present in 0 expression, and state_car is
1, detection terminates, state_car 0, removes a frame data, continues step 1 and arrives step 8.
2. the Parking quick determination method according to claim 1 based on HEVC code streams, it is characterised in that step 2 institute
State and configure virtual motion vector for each intraframe coding unit, comprise the following steps:
Step 1:Initialization frame interior coding unit failure allocation result counts mark Intra_flag=0, and according to intraframe coding
Dividing elements size is divided into following grade:
<mrow>
<mi>I</mi>
<mi>n</mi>
<mi>t</mi>
<mi>r</mi>
<mi>a</mi>
<mo>_</mo>
<mi>l</mi>
<mi>e</mi>
<mi>v</mi>
<mi>e</mi>
<mi>l</mi>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
<mtd>
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mn>64</mn>
<mo>&times;</mo>
<mn>64</mn>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>2</mn>
</mtd>
<mtd>
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mn>32</mn>
<mo>&times;</mo>
<mn>32</mn>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>3</mn>
</mtd>
<mtd>
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mn>16</mn>
<mo>&times;</mo>
<mn>16</mn>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>4</mn>
</mtd>
<mtd>
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mn>8</mn>
<mo>&times;</mo>
<mn>8</mn>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>5</mn>
</mtd>
<mtd>
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<mn>4</mn>
<mo>&times;</mo>
<mn>4</mn>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
The intraframe coding unit that note does not configure motion vector is BnoMV;
Step 2:Travel through BnoMV, and the prediction residual res of intraframe coding unit is calculated, specific computational methods are as follows:
(1) the residual error data r of pixel in current intraframe coding unit is readi k, wherein k represent current coded unit entirely drawing
The index ID of subdivision, it is integer, and i expression current pixel points are 1 in the index ID, i of current intraframe coding unit size
To current coded unit size c_size;
(2) residual error of current coded unit is calculatedWherein
Step 3:Virtual motion vector is distributed to intraframe coding unit, specific distribution method is as follows:
(1) if BnoMVIntra_level be less than or equal to ThB, then by BnoMVIt is considered as background environment part, while will corresponding BnoMV
Motion vector be set to (0,0);
(2) if BnoMVIntra_level be more than or equal to ThFOr BnoMVResidual error res be more than threshold value Thres, then by BnoMVDepending on
For moving vehicle part, meanwhile, extract BnoMVThe motion vector of surrounding interframe encode unit, non-zero motion vectors are sat using pole
Mark median filtering method extracts median motion vector, and current B is assigned with thisnoMVBlock, if adjacent encoder unit is all to use frame
The motion vector of interior coding pattern or neighboring prediction coding unit is zero, then retains current intraframe coding unit mode, temporarily
Do not configure virtual motion vector;
(3) if BnoMVIntra_level be more than ThBAnd it is less than ThF, while res is less than Thres, then B is extractednoMVSurrounding is adjacent
The motion vector of interframe encode unit, median motion vector is extracted using polar coordinates median filtering method to motion vector, assigned
Current BnoMVBlock, if adjacent encoder unit is all intra-frame encoding mode, retain current intraframe coding unit mode, Zan Bupei
Put virtual motion vector;
Step 4:Judge whether the distribution of intraframe coding unit motion vector succeeds, Intra_flag=is made if distribution is failed
Intra_flag+1;Otherwise, next step is performed;
Step 5:Judge whether current intraframe coding unit travels through completion, if current intraframe coding unit does not travel through completion,
Jump procedure 2;Otherwise, next step is performed;
Step 6:If Intra_flag > 0, jump procedure 1;Otherwise circulation is terminated.
3. the Parking quick determination method according to claim 2 based on HEVC code streams, it is characterised in that described to make
Median motion vector is extracted with polar coordinates median filtering method to comprise the steps of:
Step 1:By the unit of the inter prediction encoding unit universal formulation in the current neighborhood of intraframe coding unit 4 to 4 × 4 sizes
Grade, and the motion vector after division is allocated according to the motion vector of original predictive unit;
Step 2:Motion vector after division is saved in internal memory, asks for the angle, θ of all motion vectorsi, range value | | vi|
|2, and respectively by both by sequence from small to large;
Step 3:Angle, θ is asked for using medium filteringiAnd range value | | vi||2Intermediate value, and using the median vector as present frame
The motion vector of interior coding unit.
4. the Parking quick determination method according to claim 1 based on HEVC code streams, it is characterised in that step 3 institute
State and motion vector is handled, form motion vector field, comprise the following steps:
Step 1:The processing of motion vector Spatial normalization, different size coding unit and corresponding motion vector are divided into unification
The calibrated bolck and respective motion vectors of size 4 × 4, method are as follows:It is more than 4 × 4 part for coding unit size, this is compiled
Code unit is divided into the calibrated bolck of 4 × 4 sizes, and the motion vector of all calibrated bolcks is the coding unit motion vector before division;
Step 2:The processing of motion vector temporal criterionization, the motion vector based on multi-reference frame extracted in HEVC is converted into together
One adjacent video frames correspond to the motion vector of calibrated bolck, it is assumed that current video P frames are designated as Ic, IcIn the i-th predictive coding unit
It is designated as Bi, predictive coding unit BiMotion vector be designated as MV (Bi), its reference frame is designated as Ir, subscript c, i, r are integer number here
Word identifies, the motion vector of temporal criterion
5. the Parking quick determination method according to claim 1 based on HEVC code streams, it is characterised in that step 4 institute
The motion vector field denoising stated, comprises the following steps:
Step 1:Calculate motion vector MV horizontal MVxWith longitudinal component MVyAbsolute value sum SumMV, according to SumMVWith threshold value
T comparison decision MV value, method are as follows:
<mrow>
<mi>M</mi>
<mi>V</mi>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mi>M</mi>
<mi>V</mi>
</mrow>
</mtd>
<mtd>
<mrow>
<mrow>
<mo>|</mo>
<mrow>
<msub>
<mi>MV</mi>
<mi>x</mi>
</msub>
</mrow>
<mo>|</mo>
</mrow>
<mo>+</mo>
<mrow>
<mo>|</mo>
<mrow>
<msub>
<mi>MV</mi>
<mi>y</mi>
</msub>
</mrow>
<mo>|</mo>
</mrow>
<mo>&GreaterEqual;</mo>
<mi>T</mi>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
<mtd>
<mrow>
<mrow>
<mo>|</mo>
<mrow>
<msub>
<mi>MV</mi>
<mi>x</mi>
</msub>
</mrow>
<mo>|</mo>
</mrow>
<mo>+</mo>
<mrow>
<mo>|</mo>
<mrow>
<msub>
<mi>MV</mi>
<mi>y</mi>
</msub>
</mrow>
<mo>|</mo>
</mrow>
<mo><</mo>
<mi>T</mi>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
T selection is relevant with the motion estimation techniques used in HEVC video encoding standards;
Step 2:Using 3 × 3 templates pairMotion vector carry out median filter process.
6. the Parking quick determination method according to claim 1 based on HEVC code streams, it is characterised in that calculate step
Region minimum enclosed rectangle described in rapid 8 includes calibrated bolck number N umtotalMethod be:
Step 1:Ask for motion vector fieldMiddle flag_end flags are 1 calibrated bolck number N umstop;
Step 2:Ask for including the minimum enclosed rectangle rect for the calibrated bolck that flag_end flags are 1;
Step 3:Ask for the number N um for all calibrated bolcks that rect is includedtotal。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710367626.XA CN107404653B (en) | 2017-05-23 | 2017-05-23 | A kind of Parking rapid detection method of HEVC code stream |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710367626.XA CN107404653B (en) | 2017-05-23 | 2017-05-23 | A kind of Parking rapid detection method of HEVC code stream |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107404653A true CN107404653A (en) | 2017-11-28 |
CN107404653B CN107404653B (en) | 2019-10-18 |
Family
ID=60404717
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710367626.XA Active CN107404653B (en) | 2017-05-23 | 2017-05-23 | A kind of Parking rapid detection method of HEVC code stream |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107404653B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107742313A (en) * | 2017-09-26 | 2018-02-27 | 华勤通讯技术有限公司 | A kind of data compression method and equipment applied to vector space |
CN108256511A (en) * | 2018-03-15 | 2018-07-06 | 太原理工大学 | Body movement detection method based on Video coding code stream |
CN110163107A (en) * | 2019-04-22 | 2019-08-23 | 智慧互通科技有限公司 | A kind of method and device based on video frame identification Roadside Parking behavior |
CN112714316A (en) * | 2020-12-21 | 2021-04-27 | 太原智林信息技术股份有限公司 | Regular mark detection and classification identification method based on video code stream |
CN113014925A (en) * | 2021-01-27 | 2021-06-22 | 重庆邮电大学 | H.266/VVC inter-frame coding CU fast dividing method and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103491334A (en) * | 2013-09-11 | 2014-01-01 | 浙江大学 | Video transcode method from H264 to HEVC based on region feature analysis |
CN104301735A (en) * | 2014-10-31 | 2015-01-21 | 武汉大学 | Method and system for global encoding of urban traffic surveillance video |
CN104599502A (en) * | 2015-02-13 | 2015-05-06 | 重庆邮电大学 | Method for traffic flow statistics based on video monitoring |
CN105409222A (en) * | 2013-07-17 | 2016-03-16 | 索尼公司 | Image processing device and method |
CN106340031A (en) * | 2015-07-07 | 2017-01-18 | 香港生产力促进局 | Method and device for detecting moving object |
-
2017
- 2017-05-23 CN CN201710367626.XA patent/CN107404653B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105409222A (en) * | 2013-07-17 | 2016-03-16 | 索尼公司 | Image processing device and method |
CN103491334A (en) * | 2013-09-11 | 2014-01-01 | 浙江大学 | Video transcode method from H264 to HEVC based on region feature analysis |
CN104301735A (en) * | 2014-10-31 | 2015-01-21 | 武汉大学 | Method and system for global encoding of urban traffic surveillance video |
CN104599502A (en) * | 2015-02-13 | 2015-05-06 | 重庆邮电大学 | Method for traffic flow statistics based on video monitoring |
CN106340031A (en) * | 2015-07-07 | 2017-01-18 | 香港生产力促进局 | Method and device for detecting moving object |
Non-Patent Citations (4)
Title |
---|
LIANG ZHAO等: "HEVC Compressed Domain Moving Object Detection and Classfication", 《IEEE EXPLORE》 * |
吕小军: "面向新一代移动通信的视频监控***研究与设计", 《中国优秀硕士学位论文全文数据库》 * |
宋耀: "交通监控视频中的车辆异常行为检测", 《中国优秀硕士学位论文全文数据库》 * |
时秀波: "适于交通监控的内容自适应空间可伸缩视频编码算法", 《计算机应用与软件》 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107742313A (en) * | 2017-09-26 | 2018-02-27 | 华勤通讯技术有限公司 | A kind of data compression method and equipment applied to vector space |
CN107742313B (en) * | 2017-09-26 | 2021-02-09 | 华勤技术股份有限公司 | Data compression method and device applied to vector space |
CN108256511A (en) * | 2018-03-15 | 2018-07-06 | 太原理工大学 | Body movement detection method based on Video coding code stream |
CN108256511B (en) * | 2018-03-15 | 2022-03-29 | 太原理工大学 | Human motion detection method based on video coding code stream |
CN110163107A (en) * | 2019-04-22 | 2019-08-23 | 智慧互通科技有限公司 | A kind of method and device based on video frame identification Roadside Parking behavior |
CN112714316A (en) * | 2020-12-21 | 2021-04-27 | 太原智林信息技术股份有限公司 | Regular mark detection and classification identification method based on video code stream |
CN113014925A (en) * | 2021-01-27 | 2021-06-22 | 重庆邮电大学 | H.266/VVC inter-frame coding CU fast dividing method and storage medium |
CN113014925B (en) * | 2021-01-27 | 2022-08-23 | 重庆邮电大学 | H.266/VVC inter-frame coding CU fast dividing method and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN107404653B (en) | 2019-10-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107404653A (en) | A kind of Parking quick determination method of HEVC code streams | |
TWI445411B (en) | Method for performing local motion vector derivation during video coding of a coding unit, and associated apparatus | |
RU2708347C1 (en) | Image encoding method and device and image decoding method and device | |
CN104268900B (en) | Motion object detection method and device | |
CN106713935A (en) | Fast method for HEVC (High Efficiency Video Coding) block size partition based on Bayes decision | |
KR101822687B1 (en) | Method for reconstructing a current block of an image and corresponding encoding method, corresponding devices as well as storage medium carrying an images encoded in a bit stream | |
CN104869421B (en) | Saliency detection method based on overall motion estimation | |
CN107396102A (en) | A kind of inter-frame mode fast selecting method and device based on Merge technological movement vectors | |
KR102261669B1 (en) | Artificial Neural Network Based Object Region Detection Method, Device and Computer Program Thereof | |
CN107087200A (en) | Coding mode advance decision method is skipped for high efficiency video encoding standard | |
CN103826125B (en) | Concentration analysis method and device for compression monitor video | |
CN103856781B (en) | Self-adaptation threshold value video streaming multi-texture-direction error concealment method | |
CN101266132A (en) | Running disorder detection method based on MPFG movement vector | |
CN108347612A (en) | A kind of monitored video compression and reconstructing method of view-based access control model attention mechanism | |
Zhang et al. | Quadtree decision for depth intra coding in 3D-HEVC by good feature | |
CN108989799A (en) | A kind of selection method, device and the electronic equipment of coding unit reference frame | |
CN106791828A (en) | High performance video code-transferring method and its transcoder based on machine learning | |
Shi et al. | Reinforced bit allocation under task-driven semantic distortion metrics | |
CN113392726B (en) | Method, system, terminal and medium for identifying and detecting head of person in outdoor monitoring scene | |
Laumer et al. | Compressed domain moving object detection by spatio-temporal analysis of H. 264/AVC syntax elements | |
CN105915916A (en) | Video steganalysis method based on motion vector rate-distortion performance estimation | |
CN110149512A (en) | Inter-prediction accelerated method, control device, electronic device, computer storage medium and equipment | |
CN106131553A (en) | A kind of video steganalysis method based on motion vector residual error dependency | |
CN104104948B (en) | Video transcoding method and video code translator | |
CN105791863B (en) | 3D-HEVC depth map intra-frame predictive encoding method based on layer |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |