CN102378012A - Data hiding-based H.264 video transmission error code recovery method - Google Patents
Data hiding-based H.264 video transmission error code recovery method Download PDFInfo
- Publication number
- CN102378012A CN102378012A CN2011103823172A CN201110382317A CN102378012A CN 102378012 A CN102378012 A CN 102378012A CN 2011103823172 A CN2011103823172 A CN 2011103823172A CN 201110382317 A CN201110382317 A CN 201110382317A CN 102378012 A CN102378012 A CN 102378012A
- Authority
- CN
- China
- Prior art keywords
- frame
- error code
- macro block
- data
- video
- 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
Links
Images
Abstract
The invention provides a data hiding-based H.264 video transmission frame I error code recovery method by aiming at how to improve H.264 video transmission error code recovery effect and improve the quality of wireless video transmission. The method comprises the following steps of: 1) side information of video frames is extracted in a self-adaptive way at a coding end by using the data hiding-based method and scene switching information and is hidden in a video streaming group, so that the side information can be extracted at the coding end, the video transmission error code recovery can be realized by using the self-adaptive algorithm, the quality of the H.264 video transmission in various packet loss environments is improved, and especially, the method has a better error code recovery effect on scene switching and large movement pictures; and 2) data hiding is realized by using an extended field of a real time protocol (RTP ) message header, so that the problem of insufficient important data hiding spaces for video frames is better solved, the original video coding data is not damaged, and the quality of the video reconstruction is improved.
Description
Technical field
The invention belongs to the wireless video communication field, relate in particular in a kind of wireless video communication H.264 video transmission error code restoration methods based on data hidden.
Background technology
The transmission channel of cordless communication network and the Internet is also unreliable; In the transmission course of video bit stream; Problem such as channel disturbance, network congestion will cause phenomenons such as random bit mistake, sudden mistake and packet loss error code, causes the video data information that is transmitted to lose.Because the effect of its motion compensation mechanism, the error code in the predictive coding will spread on time and space to the H.264 standard of low-bit rate video communications applications, and the mistake in any present frame all will influence the subsequent frame with reference to it; Error code in the variable-length encoding sequence influences very big zone in the GOB with spatial diffusion, up to next synchronous points; This series standard adopts the entropy coding mode with the acquisition high compression rate, and the greatest drawback of entropy coding is very responsive to error code.Even a small amount of error code appears in code stream in transmission course, all possibly cause the visual quality severe exacerbation.Therefore, H.264 the transmission of the robustness of video flowing is used very important for radio multimedium.
H.264 video transmission error code restoration methods basic thought based on data hidden is to extract the significant data that auxiliary error code recovers at coding side, and is hidden in the code stream, extracts the recovery that the significant data of hiding is used for the video transmission error code in decoding end.This method has more available information with respect to the method for shielding error code of no Information hiding, the better quality of reconstruction video image; And, saved channel resource and can obtain visual quality preferably also, thereby be more suitable for channel resource anxiety, wireless transmission that packet loss is high with respect to adding redundant video flowing guard method.Simultaneously, this method has also reduced the computation complexity of decoder.
Video transmission error code restoration methods based on data hidden is mainly concerned with 4 problems: how to extract significant data, how to select the hiding place, how to hide, how to utilize significant data to carry out the error code recovery.Present method all is to study under the specific video transmitting scene around these 4 problems.And how to select suitable error code recovery algorithms, how to select hidden method not to be well solved at present.Error code restoration methods based on data hidden is at present: at first dominate edge direction as significant data at the interior pixels or the macro block of coding side extraction I frame macro block; Adopt the odd and even data embedding grammar significant data of macro block to be embedded in the adjacent macroblocks of same frame then or in the DCT coefficient of next P frame macro block; If then extracting corresponding significant data, the existing error code of the I-frame video of reception outflow adopt BI (bilinear interpolation) algorithm or DI (directional interpolation) algorithm to carry out the error code recovery in decoding end at last.
The shortcoming of existing method mainly contains: 1) these class methods are not distinguished the I frame type, all extract identical to all I frames
The significant data of type, and only adopt single error code recovery algorithms to carry out error code and recover, thereby cause the error code recovery effects not good.2) some significant data being hidden into down a slice is very easy to cause initial data and hiding data to lose simultaneously.3) adopt the odd and even data embedding grammar to carry out the original coding data that data hidden can damage the I frame, the original coding data of especially when the hiding data amount is very big, destroying are more.4) the non-zero quantization DCT coefficient is less in many scenes, and the hiding data amount situation that significant data can't be hidden will occur more for a long time.
Summary of the invention
Technical problem:The present invention proposes a kind of H.264 video transmission error code restoration methods based on data hidden; I frame error code recovery algorithms through the design space-time adaptive has further improved the wireless video transmission error code recovery effects under the high packet loss; As hiding the place, solved the hiding problem of significant data through design Real-time Transport Protocol message extended head.
Technical scheme:The inventive method is extracted the various significant datas of I frame macro block adaptively at coding side; And the significant data that extracts is hidden into the extended head of next frame RTP message; Extract significant data and adopt the space-time adaptive algorithm that the I frame is carried out the error code recovery in decoding end then, improved the error code recovery effects of I frame.
This method comprises:
Process one: the self adaptation significant data based on the scene handover information extracts and hidden method: coding side carries out scene detection to the I frame, and the I frame is divided into 3 types; Dissimilar according to the I frame then, the significant data of coding side extracted in self-adaptive frame of video, and be hidden into the extended head of next frame realtime transmission protocol RTP message;
Process two: based on the self adaptation error code restoration methods of hiding data: decoding end if there is error code, is then at first extracted the significant data of hiding to rebuilding the I frame, adopts different algorithms to carry out error code according to frame type then and recovers.
Extract and hidden method is based on the self adaptation significant data of scene handover information: coding side carries out scene detection to the I frame, and the I frame is divided into 3 types of big motion picture, little motion picture and scene switch frames; Dissimilar according to the I frame, coding side then extracts different significant datas; If the I frame is little motion picture, then extract the sign of indication time domain recovery algorithms; If the I frame is big motion picture, then extract the significant data of different spatial domains recovery algorithms according to the intra-frame macro block type self adaption; If macro block is smooth macro block, complex texture macro block, then extract the brightness value of 4 pixels of interior macroblocks; If macro block is the simple textures macro block, then extract the leading edge direction of macro block; If the I frame is the scene switch frame, then add 2/3 I frame redundant data protection; For the non-scene change I frame of video sequence, realize that adaptive significant data is hiding in the 0th RTP bag of first P frame of coding side behind each I frame then; If the I frame is little motion picture, then hiding data is that little motion picture indicates, and the expression decoding end adopts the time domain concealing method; If the I frame is big motion picture, then hide big motion picture unlabeled data, the expression decoding end adopts the spatial concealment method; Then hide directional interpolation algorithm DI significant data and BI algorithm significant data adaptively according to macro block (mb) type; If smooth macro block, complex texture macro block, the then brightness value of hiding 4 pixels of interior macroblocks; If simple textures macro block, the then leading edge direction of hiding macro block.
Self adaptation error code restoration methods based on hiding data is: when H.264 decoding end receives code stream, at first code stream is carried out preliminary treatment, remove redundant data packets, rebuild this frame of video then; There is error code if rebuild the I frame, then at first extracts the significant data of hiding; If loss of vital data judges then whether the I frame data are lost more than 60%, recover the weighted pixel average algorithm WPA interpolation reconstruction error code that then adopts JM to carry if not if then adopt the best neighborhood matching algorithm BNM of time domain to carry out error code; If significant data is not lost, then utilize significant data to judge whether the I frame is big motion picture; If the I frame is big motion picture; Then the type self adaption ground according to the error code macro block adopts spatial domain code error shielding algorithm to recover error code; If the simple textures macro block then adopts directional interpolation algorithm DI according to the macro block dominant direction, if smooth macro block and complex texture macro block then adopt bilinear interpolation algorithm BI according to 4 intermediate luminance pixel values; If the I frame is little motion picture, then adopt the BNM algorithm of time domain to carry out the error code recovery; For rebuilding the P frame, if there is error code, the boundary matching distortion criterion SMD that then directly uses JM (Joint Model) to carry carries out error code and recovers.
Beneficial effect:
1) based on the method for data hidden and scene handover information side information in coding side extracted in self-adaptive frame of video; And be hidden in the stream of video packets; Thereby can extract side information in decoding end adopts adaptive algorithm to realize that the video transmission error code recovers; Improved under the various packet loss environment H.264 video transmission quality, particularly scene being switched with big motion picture has error code recovery effects preferably.
2) through utilizing the extended field of RTP header, realize data hidden, thereby solved the not enough problem of frame of video significant data concealed space preferably, do not destroy the original video coded data simultaneously, improved the video reconstruction quality.
Description of drawings
Fig. 1 is that I frame significant data of the present invention extracts flow chart;
Fig. 2 is the error code restoration methods of decoding end of the present invention based on data hidden;
Fig. 3 is to use the news of the inventive method and existing method and football mixed video simulation performance to compare.Wherein: (a) the PSNR curve under 5% packet loss, (b) the PSNR curve under 10% packet loss, (c) the PSNR curve under 20% packet loss.
Embodiment
This method mainly comprises four parts: flexible macro-block ordering FMO, the significant data of coding side extracted in self-adaptive macro block, data hidden, each algorithm of decoding end utilize significant data to carry out the recovery of self adaptation error code.
At first, coding side adopts self-defining checker board pattern that frame of video is divided into 3.Then, coding side carries out scene detection to the I frame, and the I frame is divided into 3 types: big motion picture, little motion picture and scene switch frame.Dissimilar according to the I frame, coding side then extracts different significant datas.If the I frame is little motion picture, then extract the significant data of time domain recovery algorithms.If the I frame is big motion picture, be the significant data that smooth macro block, simple textures macro block and complex texture macro block adaptive extract different spatial domains recovery algorithms then according to macro block (mb) type.If the I frame is the scene switch frame, then add 2/3 I frame redundant data protection.For the non-scene change I frame of video sequence, realize that adaptive significant data is hiding in the 0th RTP bag of first P frame of coding side behind each I frame then.If the I frame is little motion picture, then hiding data is that little motion picture indicates, and the expression decoding end adopts the time domain concealing method.If the I frame is big motion picture, then hide big motion picture unlabeled data, the expression decoding end adopts the spatial concealment method.Be smooth macro block according to macro block (mb) type then, simple textures macro block and complex texture macro block adaptive ground hide DI algorithm significant data and BI algorithm significant data.
At last, when H.264 decoding end receives code stream, at first code stream is carried out preliminary treatment, remove redundant data packets, rebuild this frame of video then.If there is error code in this reconstruction frames, then adopts adaptive approach to select different algorithms to extract required important information and carry out the error code recovery, if do not have error code, the decoder buffer of then writing direct.
For rebuilding the I frame,, then at first extract the significant data of hiding if there is error code.If loss of vital data judges then whether the I frame data are lost more than 60%, recover if then adopt the BNM algorithm of time domain to carry out error code, then adopt if not
WPA (weighted pixel is average) the algorithm interpolation reconstruction error code that carries with JM.If significant data is not lost, then utilize significant data to judge whether the I frame is big motion picture.If the I frame is big motion picture; Then the type self adaption ground according to the error code macro block adopts spatial domain code error shielding algorithm to recover error code; If the simple textures macro block then adopts DI (directional interpolation) algorithm according to the macro block dominant direction, if smooth macro block and complex texture macro block then adopt BI (bilinear interpolation) algorithm according to 4 intermediate luminance pixel values.If the I frame is little motion picture, then adopt the BNM algorithm of time domain to carry out the error code recovery.For rebuilding the P frame, if there is error code, the SMD criterion that then directly carries with JM is carried out error code and is recovered.
Instance:
1.FMO
The present invention at first uses flexible macro-block ordering (FMO) technology H.264.Here adopt self-defining checker board pattern that a QCIF frame of video is divided into 3 sheets (slice), promptly each sheet comprises 33 macro blocks.
2. extraction significant data
For the I frame in the video sequence, adopt the method shown in the accompanying drawing 1 to extract significant data at coding side.
(1) scene detection
At coding side the I frame is adopted scene-detection algorithms, judge that according to MAD (formula (1)) method present frame and previous frame have or not scene change.If
is greater than threshold value T3 (20≤T3≤40), then occurrence scene conversion.If
then is big motion picture less than threshold value T3 and greater than threshold value T (5≤T≤15), with " 1 " expression; Otherwise be little motion picture,
With " 0 " expression.
（1）
Wherein, The pixel value of
and
expression current I frame and previous P frame, total number of pixels of
expression one frame.
(2) BI (bilinear interpolation) algorithm extracts significant data
Look for four pixels at the middle part of a macro block, its brightness value is extracted as significant data at coding side, adopt the BI algorithm to carry out the error code recovery in decoding end then losing macro block.
(3) DI (directional interpolation) algorithm extracts significant data
Suppose in the macro block that a certain pixel is
, the amplitude
and the direction value
of marginal point in using sobel edge detection operator can calculating
.
At first judge whether there is edge pixel in the macro block according to
, as long as the normalization gradient absolute value of pixel is exactly an edge pixel greater than pre-set threshold T1 (0.2≤T1≤0.4).The macro block that has edge pixel classifies as the edge macro block, otherwise classifies as smooth macro block.If edge macro block; Then the gradient direction
with each pixel in the macro block is quantified as 8 directions (interval 22.5 degree; 0 represents 0 degree, and 1 represents 22.5 degree ... 7 represent 157.5 degree).Calculate the variance
that each macro block quantizes gradient direction then; Through comparing, the edge macro block further is divided into the macro block of simple textures macro block and complex texture with preset threshold value T2 (0.3≤T2≤0.5).
At last, be that smooth macro block, simple textures macro block and complex texture macro block extract different significant datas according to macro block (mb) type.If smooth macro block then uses " 9 " to indicate, coding 4bit.If the simple textures macro block then belongs to this macro block all gradient absolute value additions of each gradient direction (0-7) earlier, select then and the dominant direction (0-7 represent 8 directions) of maximum gradient direction as macro block, coding 4bits.If the complex texture macro block, then macro block further is subdivided into 48 * 8 piece, calculate the dominant direction (0-7 representes 8 directions) of each piece respectively.
(4) significant data of BNM algorithm is directly represented with little motion picture identifier " 0 ".
3. data hidden
Real-time Transport Protocol is the agreement of the suitable real-time Data Transmission of IETF proposition; Be supported in transmitting real-time data in single goal broadcasting and the service of multiple target radio network; The transmission service of point-to-point can be provided for the real-time media data, be at present H.264 video flowing transmission use agreement more widely.The present invention utilizes the extended head of Real-time Transport Protocol to carry out data hidden.
Non-scene change I frame for video sequence:
Realize that adaptive significant data is hiding in the 0th RTP bag of first P frame of coding side behind each I frame.
(I) if the I frame is little motion picture IframeChangeFlag=0, only need the extended head of 4bytes, wherein self-defining 2 byte field are " 0 ", the expression decoding end adopts the time domain concealing method.
(II) if the I frame is big motion picture IframeChangeFlag=1, fixedly self-defining 2 byte field of extended head are " 1 ", the expression decoding end adopts the spatial concealment method.According to smooth macro block, simple textures macro block and complex texture macro block are hidden DI algorithm significant data and BI algorithm significant data adaptively at extended head then.
4. error code recovers
When H.264 decoding end receives code stream, at first code stream is carried out preliminary treatment, remove redundant data packets, rebuild this two field picture then.If there is error code in this reconstruction frames, then adopt the adaptive approach of accompanying drawing 2 to select the required important information of different algorithms extraction to carry out the error code recovery, if do not have error code, the decoder buffer of then writing direct.
For rebuilding the I frame:, then at first extract the significant data of hiding if there is error code.If loss of vital data judges then whether the I frame data are lost more than 60%, recover WPA (weighted pixel is average) the algorithm interpolation reconstruction error code that then adopts JM to carry if not if then adopt the BNM algorithm of time domain to carry out error code.If significant data is not lost, then utilize significant data to judge whether the I frame is big motion picture.If the I frame is big motion picture; Then the type self adaption ground according to the error code macro block adopts spatial domain code error shielding algorithm to recover error code; If the simple textures macro block then adopts DI (directional interpolation) algorithm according to the macro block dominant direction, if smooth macro block and complex texture macro block then adopt BI (bilinear interpolation) algorithm according to 4 intermediate luminance pixel values.If the I frame is little motion picture, then adopt the BNM algorithm of time domain to carry out the error code recovery.
For rebuilding the P frame: if there is error code, the SMD criterion that then directly carries with JM is carried out error code and is recovered.
Simulation result of the present invention
Reference software JM8.6 H.264 is as the emulation experiment platform for the present invention's employing, and adopts the C language to realize.Adopt H.264 basic class that 200 frame video sequences of a plurality of standard QCIF forms are carried out the encoding and decoding test in the emulation.Wherein, the size of GOP is set to 10, and frame per second was 30 frame/seconds; Quantization parameter is 28, and reference frame number S is set at 3, and FMO is divided into 3 sheets; Each sheet is made up of 33 macro blocks, video packets by one completely sheet form, if transmission course generation mistake then whole video Bao Jun lose.The T value is 8, and the T1 value is 0.3, and the T2 value is 0.3, and the T3 value is 22.Hunting zone H is 32 in the BNM algorithm, and L is 32, and development length m is 2.Emulation packet loss model is packet loss model at random.
The effect of error code recovery algorithms is general relevant with concrete video sequence, and regular degree and severe that different video sequence texture complexities is different, move in the scene also differ greatly.Accompanying drawing 3 be news and the PSNR curve ratio of football mixed video under the different packet loss rate.Say that from overall performance the I frame adaptive error code recovery algorithms performance that the present invention is based on data hidden is superior to the spatial domain error code recovery algorithms of remaining data hidden.The error code that the DH-DI algorithm relatively is fit to the simple edges macro block recovers, and the spatial domain error code that the DH-BI algorithm relatively is fit to level and smooth macro block and complex texture macro block recovers, and algorithm utilization judgement scene change of the present invention has been introduced space-time adaptive error code restoration methods.Can find out that from accompanying drawing 3 DH-DI algorithm, DH-BI algorithm, DH-BI DI algorithm are very little with respect to the WPA algorithms to improve, and the relative WPA algorithm of the inventive method has greatly improved, and loses for 200 frame news and football mixed video 10%
Mean P SNR has the improvement of 3.2db during bag, and mean P SNR has the improvement of 3db during 20% packet loss.In addition, the present invention has also reduced the hiding data amount.
Claims (3)
1. H.264 video transmission error code restoration methods based on data hidden is characterized in that this method comprises:
Process one: the self adaptation significant data based on the scene handover information extracts and hidden method: coding side carries out scene detection to the I frame, and the I frame is divided into 3 types; Dissimilar according to the I frame then, the significant data of coding side extracted in self-adaptive frame of video, and be hidden into the extended head of next frame realtime transmission protocol RTP message;
Process two: based on the self adaptation error code restoration methods of hiding data: decoding end if there is error code, is then at first extracted the significant data of hiding to rebuilding the I frame, adopts different algorithms to carry out error code according to frame type then and recovers.
2. the H.264 video transmission error code restoration methods based on data hidden as claimed in claim 1; It is characterized in that extracting and hidden method is based on the self adaptation significant data of scene handover information: coding side carries out scene detection to the I frame, and the I frame is divided into 3 types of big motion picture, little motion picture and scene switch frames; Dissimilar according to the I frame, coding side then extracts different significant datas; If the I frame is little motion picture, then extract the sign of indication time domain recovery algorithms; If the I frame is big motion picture, then extract the significant data of different spatial domains recovery algorithms according to the intra-frame macro block type self adaption; If macro block is smooth macro block, complex texture macro block, then extract the brightness value of 4 pixels of interior macroblocks; If macro block is the simple textures macro block, then extract the leading edge direction of macro block; If the I frame is the scene switch frame, then add 2/3 I frame redundant data protection; For the non-scene change I frame of video sequence, realize that adaptive significant data is hiding in the 0th RTP bag of first P frame of coding side behind each I frame then; If the I frame is little motion picture, then hiding data is that little motion picture indicates, and the expression decoding end adopts the time domain concealing method; If the I frame is big motion picture, then hide big motion picture unlabeled data, the expression decoding end adopts the spatial concealment method; Then hide directional interpolation algorithm DI significant data and BI algorithm significant data adaptively according to macro block (mb) type; If smooth macro block, complex texture macro block, the then brightness value of hiding 4 pixels of interior macroblocks; If simple textures macro block, the then leading edge direction of hiding macro block.
3. the H.264 video transmission error code restoration methods based on data hidden as claimed in claim 1; It is characterized in that based on the self adaptation error code restoration methods of hiding data being: when H.264 decoding end receives code stream; At first code stream is carried out preliminary treatment; Remove redundant data packets, rebuild this frame of video then; There is error code if rebuild the I frame, then at first extracts the significant data of hiding; If loss of vital data judges then whether the I frame data are lost more than 60%, recover the weighted pixel average algorithm WPA interpolation reconstruction error code that then adopts JM to carry if not if then adopt the best neighborhood matching algorithm BNM of time domain to carry out error code; If significant data is not lost, then utilize significant data to judge whether the I frame is big motion picture; If the I frame is big motion picture; Then the type self adaption ground according to the error code macro block adopts spatial domain code error shielding algorithm to recover error code; If the simple textures macro block then adopts directional interpolation algorithm DI according to the macro block dominant direction, if smooth macro block and complex texture macro block then adopt bilinear interpolation algorithm BI according to 4 intermediate luminance pixel values; If the I frame is little motion picture, then adopt the BNM algorithm of time domain to carry out the error code recovery; For rebuilding the P frame, if there is error code, the boundary matching distortion criterion SMD that then directly uses JM (Joint Model) to carry carries out error code and recovers.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011103823172A CN102378012A (en) | 2011-11-26 | 2011-11-26 | Data hiding-based H.264 video transmission error code recovery method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011103823172A CN102378012A (en) | 2011-11-26 | 2011-11-26 | Data hiding-based H.264 video transmission error code recovery method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102378012A true CN102378012A (en) | 2012-03-14 |
Family
ID=45795896
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011103823172A Pending CN102378012A (en) | 2011-11-26 | 2011-11-26 | Data hiding-based H.264 video transmission error code recovery method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102378012A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103051903A (en) * | 2012-12-24 | 2013-04-17 | 四川九洲电器集团有限责任公司 | Space adaptive H.264 video I frame error concealment method |
| CN105141961A (en) * | 2015-08-03 | 2015-12-09 | 中国人民解放军信息工程大学 | Spatial data double-protocol transmission method based on video steganography |
| CN103826128B (en) * | 2014-02-25 | 2017-01-25 | 宁波工程学院 | Encryption domain H.264/AVC video reversible data hiding method |
| CN110740334A (en) * | 2019-10-18 | 2020-01-31 | 福州大学 | frame-level application layer dynamic FEC encoding method |
| US10579202B2 (en) | 2012-12-28 | 2020-03-03 | Glide Talk Ltd. | Proactively preparing to display multimedia data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1581972A (en) * | 2004-05-20 | 2005-02-16 | 复旦大学 | Error-corcealed video decoding method |
| WO2006012911A1 (en) * | 2004-08-03 | 2006-02-09 | Fg Microtec Gmbh | Dynamic optimization of wireless real-time video data flow |
| CN101527850A (en) * | 2009-04-10 | 2009-09-09 | 北京航空航天大学 | Video self-recovery watermark method based on motion estimation |
-
2011
- 2011-11-26 CN CN2011103823172A patent/CN102378012A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1581972A (en) * | 2004-05-20 | 2005-02-16 | 复旦大学 | Error-corcealed video decoding method |
| WO2006012911A1 (en) * | 2004-08-03 | 2006-02-09 | Fg Microtec Gmbh | Dynamic optimization of wireless real-time video data flow |
| CN101527850A (en) * | 2009-04-10 | 2009-09-09 | 北京航空航天大学 | Video self-recovery watermark method based on motion estimation |
Non-Patent Citations (1)
| Title |
|---|
| 陈海波，董育宁: "H.264视频流自适应I帧数据隐藏和恢复方法", 《仪器仪表学报》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103051903A (en) * | 2012-12-24 | 2013-04-17 | 四川九洲电器集团有限责任公司 | Space adaptive H.264 video I frame error concealment method |
| US10579202B2 (en) | 2012-12-28 | 2020-03-03 | Glide Talk Ltd. | Proactively preparing to display multimedia data |
| US10599280B2 (en) | 2012-12-28 | 2020-03-24 | Glide Talk Ltd. | Dual mode multimedia messaging |
| US10678393B2 (en) | 2012-12-28 | 2020-06-09 | Glide Talk Ltd. | Capturing multimedia data based on user action |
| US10739933B2 (en) | 2012-12-28 | 2020-08-11 | Glide Talk Ltd. | Reduced latency server-mediated audio-video communication |
| US11144171B2 (en) | 2012-12-28 | 2021-10-12 | Glide Talk Ltd. | Reduced latency server-mediated audio-video communication |
| CN103826128B (en) * | 2014-02-25 | 2017-01-25 | 宁波工程学院 | Encryption domain H.264/AVC video reversible data hiding method |
| CN105141961A (en) * | 2015-08-03 | 2015-12-09 | 中国人民解放军信息工程大学 | Spatial data double-protocol transmission method based on video steganography |
| CN105141961B (en) * | 2015-08-03 | 2017-12-22 | 中国人民解放军信息工程大学 | A kind of double protocol transmission methods of spatial data based on video steganography |
| CN110740334A (en) * | 2019-10-18 | 2020-01-31 | 福州大学 | frame-level application layer dynamic FEC encoding method |
| CN110740334B (en) * | 2019-10-18 | 2021-08-31 | 福州大学 | Frame-level application layer dynamic FEC encoding method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8170124B2 (en) | MPEG-4 streaming system with adaptive error concealment | |
| JP2003533954A (en) | Video coding method | |
| CN102291585A (en) | Method for error concealment in video sequences | |
| CN102026001B (en) | Method for evaluating importance of video frame based on motion information | |
| CN101163250B (en) | Boundary gradient based video stream fault tolerance method | |
| CN102378012A (en) | Data hiding-based H.264 video transmission error code recovery method | |
| CN102025993B (en) | Video transmission method and system based on H.264 | |
| CN110113610B (en) | Data transmission method and device | |
| Al-Mualla et al. | Motion field interpolation for temporal error concealment | |
| WO2005125212A1 (en) | Method for video encoding and decoding process | |
| CN101242540B (en) | Error control method for video decoder chip | |
| CN102223544B (en) | Method for processing error after detecting error in H264 video stream | |
| CN102984525A (en) | Method for hiding video code stream error | |
| Chen et al. | Spatio-temporal boundary matching algorithm for temporal error concealment | |
| CN103916714A (en) | Packet-loss resistance preprocessing and recovery method for wireless network video stream | |
| Kwon et al. | Error concealment techniques for H. 263 video transmission | |
| Brailean | Wireless multimedia utilizing MPEG-4 error resilient tools | |
| CN107835422B (en) | HEVC multi-description image coding algorithm based on significance | |
| US20040228404A1 (en) | Moving picture coding method | |
| Zhang et al. | Recursive distortion estimation for hybrid digital-analog video transmission | |
| CN1219405C (en) | Image error hidding method | |
| Carreira et al. | A robust video encoding scheme to enhance error concealment of intra frames | |
| Xu et al. | Error resilient coding based on reversible data hiding and redundant slice | |
| KR20070006006A (en) | System for digital multimedia broadcast performance improvement with watermark and error concealment | |
| KR100557118B1 (en) | Moving picture decoder and method for moving picture decoding |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120314 |