CN100375520C - Video coding and decoding method for reducing data loss in transmission of coded video signals iva channel - Google Patents
Video coding and decoding method for reducing data loss in transmission of coded video signals iva channel Download PDFInfo
- Publication number
- CN100375520C CN100375520C CNB02143736XA CN02143736A CN100375520C CN 100375520 C CN100375520 C CN 100375520C CN B02143736X A CNB02143736X A CN B02143736XA CN 02143736 A CN02143736 A CN 02143736A CN 100375520 C CN100375520 C CN 100375520C
- Authority
- CN
- China
- Prior art keywords
- image
- identifier field
- encoded
- field
- coding
- 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.)
- Expired - Lifetime
Links
- 238000000034 method Methods 0.000 title claims abstract description 79
- 230000005540 biological transmission Effects 0.000 title abstract description 25
- 230000008859 change Effects 0.000 claims description 17
- 230000006854 communication Effects 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 4
- 230000004048 modification Effects 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000006835 compression Effects 0.000 description 4
- 238000007906 compression Methods 0.000 description 4
- 230000001360 synchronised effect Effects 0.000 description 4
- 230000015556 catabolic process Effects 0.000 description 3
- 238000006731 degradation reaction Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000006735 deficit Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000008439 repair process Effects 0.000 description 2
- 229920006395 saturated elastomer Polymers 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The present invention relates to a video encoded and decoded method for reducing data loss in the transmission of encoded video signals by a channel, which permits a piece of a current image free from the influence of the data loss to be decoded under the condition of data in an encoded video image (the current image) being lost even if an image head label of the current image is lost, and the encoded video image is cut into pieces and is relevant to an encoded video signal.
Description
Technical field
The image sequence that the present invention relates to encode provides the method for the method of an image sequence that is encoded and the described image encoded of decoding, and allows the image of the described present encoding of decoding under the image impairment image head target situation of a present encoding.
The invention still further relates to the coded system and the decode system of the different step that realizes described Code And Decode method.
The present invention obtains widely applying in the vision signal that is being subjected to encoding in the environment of error of transmission by Channel Transmission.
Background technology
Current, in extensive application, use video compression, such as in professional application and the application that satisfying the needs of consumers.
Video compression is applied in the application of phone or video telephone type especially, perhaps also is used between several correspondents the voice data of transmission and exchange, and sends the data of video type simultaneously.In this application, the passband of the communication channel of transmitting video data is very limited, makes to consider that a considerable amount of video datas have the acceptable image quality of forming encoded video signal.Therefore need to use be applicable to little passband video compression standard, particularly video compression standard H.263, it is exclusively used in coding and is created in small-format vision signal with the vision signal of per second tens kilobits coding.
In the scope of the application of phone or video telephone type, vision signal is usually passed through disturbed Channel Transmission, or is subjected to the passband of described signal saturated, interrupted by transmission, during wireless transmission, this interference comprises the data degradation of the vision signal of the transmission that is received by decoder especially.For limiting the effect of these interference, H.263 standard provides the protection of the video data of forming each video signal image in the encoder one-level before by transmission channel.For this purpose, this standard recommendation is in blocks each image segmentation, and these sheets of encoding independently of one another.By this way, the data degradation that takes place in the code pattern photo in present image does not influence other sheet of described present image, and this sheet can be decoded under certain conditions.The decoding of other sheet of described present image only just may under the condition that its image leader content is not lost.
A kind of method of European patent EP 1094672 explanation, its present image that has lost (or only some sheet of present image) that allows under certain condition to decode.Really, this method allows to reappear under a stable condition and is the useful coding parameter of decoding present image, and these parameters should exist in most of image leaders.
This known method has benefited from the following fact, and promptly H.263 video standard advises that each image encoded comprises an image leader, and can be cut into and comprise a head target sheet, and each sheet leader comprises an identifier field that will be identified.This identifier field is called GFID in the grammer of this standard, itself and coding rule unanimity by following rule definition: identifier field is identical for the sheet of same coded image, the image head target sign of and coded image of previous current for one comprises the sign of the identifier field of these two images, comprises the change of described present image identifier field with respect to the change of last image encoded at the image leader of the image of a present encoding.
In when decoding, if previous different, can infer that then the image leader changes, that is form described image head target field and change with the identifier field of present image, but which specific fields change of there is no telling.And according to standard H.263, the other coding rule of this method suggestion definition GFID field changes and is inferred as the new value of these fields that this current image will use at the image leader especially to allow decision which field when the decoding.
This known method has certain limitation.
At first, when with previous and identifier field that present image is related not simultaneously, this known method once only allows the change of a field in the recognition image leader.Like this, if change with the related a plurality of fields of image leader previous and that present image is related, then this method only detects the field that is considered to have limit priority, and infers that its value is the value of present image.This method does not allow to detect a plurality of fields that change in the image leader.Its result, the field of Jian Ceing may not influence the decoding of present image.
The second, when when decoding, this method suggestion is derived and the related identifier field GFID (n) of present image by the binary arithmetic operation from the identifier field GFID related with previous image (n-1), depends on the change of the field of one group of priority with increase.Obtain the value of identifier field GFID (n) then with respect to identifier field GFID (n-1).When many images are lost, (n-i) missing image for example, this method can not be worked, because consider to have preface the image of (n-1) is lost, field GFID (n-1) is unavailable when decoding.Comparison between field GFID (n) and field GFID (n-i), that is with the comparison of the related identifier field of last usable image, nonsensical because the number of the image of loss does not know that in advance this image impairment can not be planned.
At last, when when decoding,, need carry out the comparison between the identifier field GFID (n+1) of the identifier field GFID (n) of present image and next image because will determine to replace with which available image leader the image leader of present image.This means the delay of an image in equaling to decode, it can suffer damage in the video decode that must carry out visual analysis in real time.
Summary of the invention
The objective of the invention is a kind of coding of suggestion and relevant coding/decoding method, it allows to reappear the present image of decoding in the field in the image head target of losing in more firm mode under the situation that the image leader of present image is lost.
The present invention is characterised in that at first described coding method comprises the step with each of the value filling identifier field of the coding characteristic of a current image that is encoded of reflection.
Filling the value of field specifier seldom does with respect to the value of the identifier field of previous image.Really, each of identifier field directly uses the coding characteristic of the image related with them to fill.A plurality of coding characteristics are then by a given identifier field transmission.
Feature of the present invention is that also described coding/decoding method may further comprise the steps:
-identifier field of the image of the identifier field of last decoded image and present encoding is compared,
The decode image of described present encoding of the coding characteristic that-the identifier field of image by using present encoding is discerned, and the coding characteristic of being discerned by the image leader that uses described previous decoded image comes the image of described present encoding is decoded, and its condition is relatively not disclosing described identifier field:
I) image of the image of described early decoding and described present encoding is that type of prediction has different forms with them,
J) or the image of described early decoding be nonanticipating type and the image of described present encoding is a type of prediction has different forms with them.
By using the characteristic of the vision signal that produces by coding method according to the present invention, thereby can be under the situation of obliterated data during the described present image of transmission, the image head target data of particularly forming described present image, decoding are not by the image sheet of the described present image of losing influence of data.Really, by using the semanteme of described coding rule and the value in each identifier field GFID, may recover the information that comprises in the image leader of described present image, this information is not essential by the image sheet of the described present image of losing influence of data for decoding.This has such advantage, needn't use coding method for systematically use non-prediction pictures in coding/decoding method, so that the latter can carry out its task, thereby transmission channel is not by saturated.In addition, the semanteme that defines the value of each field GFID allow to improve the performance of the coding/decoding method in relating to the loss of data that can not repair decoding error.
When described coding/decoding method is in the processing of content of the identifier field GFID that explanation is filled during encoded video signal, execution is according to coding/decoding method of the present invention, and it can for example easily be realized by the Proprietary Information that sends described coding/decoding method when the transmission beginning.
The present invention has such advantage, and its keeps and operating such H.263, can predict it to it and use, even can not explain the semanteme of the vision signal that is encoded that is produced by coding method according to the present invention according to standard all coding/decoding methods H.263.
For the useful coding characteristic of decoding current encoded image reappears on the basis of the identifier field of the image of the image of present encoding relatively and previous coding.Do not introduce delay like this in decoding processing, it makes this method be used for real-time application.
The invention still further relates to coded system that realizes described coding method different step and the decode system of realizing described coding/decoding method different step.
No matter it is the Code And Decode level, lower with respect to cost the resource that in consumer product applications, can predict according to the realization of Code And Decode method of the present invention.
The invention still further relates to the communication system that comprises according to coded system of the present invention and decode system.
The invention still further relates to the portable telephone that comprises according to coded system of the present invention and decode system.
The invention still further relates to the vision signal that produces by coding method according to the present invention.
The invention still further relates to computer program for one or several steps of carrying out described Code And Decode method.
Description of drawings
By a unrestriced example, with reference to the embodiment that the following describes, these and other aspect of the present invention is with fairly obvious.
In the accompanying drawing:
Fig. 1 represents by the dissimilar image sheet of standard definition H.263,
Fig. 2 is illustrated in the main field that comprises in the image leader according to each image that is encoded of the Basic Profile file of standard H.263,
Fig. 3 is illustrated in the main field that comprises in the image leader according to each image that is encoded of the summary file of a development of standard H.263,
Fig. 4 is illustrated in the main field that comprises in the image sheet leader of GOB type,
Fig. 5 is illustrated in one arbitrarily or the main field that comprises in the image sheet leader of rectangle,
Fig. 6 explanation is according to coding/decoding method of the present invention,
Fig. 7 represents that of the present invention first uses,
Fig. 8 represents that of the present invention second uses.
Embodiment
Coding method according to the present invention comprises the step of two positions of the identifier field GFID that fills each image sheet that is encoded, the coding rule by the GFID field considering to apply by standard H.263 with by with its each related semanteme that is reflected as the used coding characteristic of this image of coding.Filling this step of GFID field can carry out during a series of uncoded video images of coding, but also can carry out according to a vision signal, it is before according to standard code H.263 and must be by a Channel Transmission, by only revising the value of these GFID fields.
Fig. 1 represents dissimilar by the image sheet of standard definition H.263.As an example, consider the image 101 of QCIF form, wherein define image sheet 102-103-104, each sheet is made up of one group of microlith with 16 * 16 sizes, such as microlith 105.
The image sheet of GOB type has the type of sheet 102, and it has and same width of image and and the same microlith highly of a microlith (or several microlith) corresponding to one group.
The image sheet of the type of piece 104 is corresponding to one group of microlith, and it has the arbitrary shape that is defined by the microlith with the connected mode assembling.
The image sheet of the type of sheet 105 is corresponding to one group of microlith, and it is shaped as rectangle, and size determines according to parameter.
When being a field when using the appendix K of standard H.263 with the SSS appointment, definition arbitrarily and rectangular shape, for arbitrary shape is got first value, for rectangle is got second value.
Cutting into image sheet provides with the encode possibility of each sheet of same image of independent mode.
Fig. 2 is illustrated in the main field that comprises in the image leader according to each image of the Basic Profile document No. of standard H.263, can not use the encoding option in this Basic Profile file.The image leader particularly including:
-PSC field, it begins code for image,
-TR field, it is the time reference that increases from the image to the image,
-PTYPE field, it is such field, it provides the information that whether has the fact of following feature about the image that is encoded especially, and described feature refers to that it can thaw and before freezed the image operated in this coding/decoding method, and about the information of picture format.
Fig. 3 is illustrated in the main field that comprises in the image leader of each image of encoding according to the summary file of the development of standard H.263, can use the encoding option in the summary file of this development, detects when it is existed by the PLUSPTYPE field.The image leader comprises the field with reference to figure 2 explanations, and the PLUSPTYPE field is made up of following field:
-field UFEP is the code word of one 3 regular length of PLUSPTYPE field, and it only is set at 111 o'clock in the 6-7-8 position of PTYPE field and just exists.When UFEP was set at " 000 ", its indication included only the PTYPE field that need use those expansions of signal indication in each image leader (MPPTYPE) in the present image leader.When being set at " 001 ", its indication comprises the PTYPE field of all expansions in the present image leader.If image type is INTRA or EI, then this field will be set at " 001 ".In addition, if having PLUSPTYPE in each in a continuous images sequence, then this field is set at " 001 ", and its frequency was by 5 seconds or 5 overtime appointments in period of images, and the two all allows a big time interval.Specifically, need overtime period UFEP=" 001 " to appear at the first image head target PLUSPTYPE field (if PLUSPTYPE exists) in each interference image, described first image has had a time durations since indication occurred from UFEP=" 001 " last time and is greater than or equal to 5 seconds, perhaps the time reference (no matter which demand all allows the long period with the time reference metric) of the 5th image after UFEP=" 001 " occurred last time.Encoder should be " 001 " by more frequent setting UFEP in being prone to wrong environment.Non-" 000 " and " 001 " value that keeps UFEP,
-OPPTYPE field, it provides about the form of image especially and is the information of the employed option of coded image,
-MPPTYPE field, its 6th is called RTYPE, and the information about the rounding type (rounding type) that realizes during the two-wire interpolation is provided, and uses described bilinear interpolation can calculate the pixel value of half.
Fig. 4 is illustrated in the main field that comprises in the image sheet leader of GOB type of each image encoded.The image sheet leader particularly including:
-GBSC field, it is a beginning code of image sheet,
-GN field, the number of a GOB of this sheet in its indicating image,
-identifier field GFID.
Fig. 5 is illustrated in arbitrarily or the main field that comprises in the image sheet leader of each image encoded of rectangle type.The image sheet leader particularly including:
-SSC field, it is the beginning code of image sheet,
-MBA field, the address of first microlith of its indication in this sheet,
-identifier field GFID.
For checking by the semantic rules of the filling identifier field GFID of H.263 standard code and provide one with the encoded video signal of operating such H.263,1 and 2 the value of field GFID must satisfy certain restriction.
Really, the value of field GFID must be identical in each leader of the piece group (if present) of same coded image or each image sheet leader.In addition, if identical in the image leader of PTYPE field in the image leader of a present image and image encoded formerly, then the value of GFID must equal not use the content of PLUSPTYPE field in these two images.Yet if the field PTYPE in the image leader of a given present image is different from the field PTYPE of the image of previous coding, the value of the GFID of described present image must be different with the value of the GFID of the image of described previous coding.Depend on field PLUSPTYPE and whether exist, the value of the GFID of present image must with image encoded formerly in the same, if field is (as by the PTYPE of standard definition H.263, PLUSPTYPE, CPFMT, EPAR, CPCFC, SSS, ELNUM, RLNUM, UUI, RPSMF, RPRP) value keep with image encoded formerly in the same, if and different, what then the value of GFID must be with the image of described previous coding is different.
The dual mode of field GFID is filled in coding method prediction according to the present invention.
-the first mode (mode 1) does not wherein change between image according to the form of the image of standard code H.263,
-the second mode (mode 2) wherein may suffer to become second form from first form between two continuous coded images according to the form of the image of standard code H.263.
Table 1 is given in the summary that picture format is disobeyed position 1 and 2 information that are filled of GFID field under the situation of image modification.
| |
|
| The |
The type of image |
| The |
Rounding type (for the image of prediction) |
| X (for non-prediction pictures) |
Table 1: disobey position 1 and 2 modes 1 of filling the GFID field under the situation of image modification at picture format and define like this:
Mode 1:
I) the position 1 reflection GFID field that will the be filled type of related image with it.For example, can be set at " 1 ", can be set at " 0 " for predictive-coded picture (being also referred to as INTER or P image) for nonanticipating coded image (being also referred to as INTRA or I image) position 1,
Ii) for predicted picture, position 2 is reflected in can be in order to affected rounding type during the bilinear interpolation of calculating one and half pixels.This is taken at the value of the PTYPE field in the related with it image leader of the GFID field that will be filled.Change the value of PTYPE between two images of predicting continuously, so that prevent the derived value of rounding and also know the image of whether losing the odd number prediction during the transmission, this is favourable.In addition, image for the current prediction of one or several nonanticipating images that continue, the PTYPE field of described present image is taken at the value of the PTYPE field of the last predicted picture before the described non-prediction pictures, so that know whether lose the nonanticipating image of odd number during the transmission
Iii) for nonanticipating image, position 2 systematically is set at " 0 " or systematically is set at " 1 ".Its value is indicated with X in table 1.
Be that a vision signal that comprises the coding of 22 consecutive images provides an example of filling the position 1 and 2 of GFID field according to mode 1 in table 2.
| |
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 1 0 | 1 1 | 1 2 | 1 3 | 1 4 | 1 5 | 1 6 | 1 7 | 1 8 | 1 9 | 2 0 | 2 1 | 2 2 | |
| Image type | P | I | P | P | P | P | I | P | P | P | P | P | P | P | P | I | I | P | P | I | I | P | |
| The |
0 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 1 | 1 | 0 | |
| The |
0 | X | 0 | 1 | 0 | 1 | |
1 | 0 | 1 | 0 | 1 | 0 | 1 | 0 | X | X | 0 | 1 | | X | 1 |
Table 2: the example of filling the position 1 and 2 of GFID field according to mode 1.
Be position 1 and 2 semantic ruleses that apply of filling the GFID field within the scope of the invention and also for satisfying by standard H.263, for mode 1 agreement:
-may must during video signal coding, revise by the option that this standard is used,
The value of-OPPTYPE field, if present, must can not be according to amending image,
The value of-SSS field is being used this field existence when wanting image encoded as the relevant appendix K of sheet with cutting, during the sign indicating number of must not being on the permanent staff according to amending image,
-under the situation that one or two continuous non-prediction pictures are provided, must change position 3-4-5.
Table 3 is given in picture format may suffer summary according to the information of the position 1 and 2 of filling the GFID field under the situation of image modification.In this case, filling mode is called mode 2.
| |
|
| The |
The type of image |
| The |
Picture format |
Table 3: the position 1 and 2 of filling the GFID field at picture format under easily according to the situation of image modification.
Mode 2:
I) the position 1 reflection GFID field that will the be filled type of related image with it.For example, can be set at " 1 ",, can be set at " 0 " for predictive-coded picture (being also referred to as INTER or P image) for nonanticipating coded image (being also referred to as INTRA or I image) position 1.So position 1 with the i of mode 1) same mode defines,
The ii) form of the image that is encoded under the 2 reflection GFID fields that will be filled of position, particularly, if picture format is in QCIF form (indicating with Q) or SQCIF (indicating with S).For example, be set at " 0 " for this position of image of SQCIF form, the image for the QCIF form is set at " 1 ", and picture format is from the 6-7-8 position definition of PTYPE field, or from the 1-2-3 position definition of OPPTYPE field.
Table 4 is that the encoded video signal that comprises 22 consecutive images provides the example of filling the position 1 and 2 of GFID according to mode 2.
| |
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 1 0 | 1 1 | 1 2 | 1 3 | 1 4 | 1 5 | 1 6 | 1 7 | 1 8 | 1 9 | 2 0 | 2 1 | 2 2 |
| Image type | P | I | P | P | P | P | I | P | P | P | P | P | P | P | P | I | P | P | P | I | P | P |
| Picture format | Q | Q | Q | Q | Q | Q | S | S | S | S | S | S | S | S | S | Q | Q | Q | Q | Q | Q | Q |
| The |
0 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 0 | 0 |
| The |
1 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
Table 4: the example of filling the position 1 and 2 of GFID field according to mode 2.
In addition, for making mode 2 observe the semantic rules that is applied to the position 1 and 2 of filling the GFID field by standard H.263:
-the option that may be used by this standard can not change when positive encoded video signal,
The value of-SSS field must can not be according to image modification during encoding, and the value of described SSS field exists when wanting image encoded as the relevant appendix K of sheet using with cutting,
-between two continuous non-prediction pictures of two images of predicting continuously and same form, the value of the position 3-4-5 of field PTYPE must be able to not change during encoded video signal,
If-use the PLUSPTYPE field, between two consecutive images of same type, the value of field UFEP and RTYPE must be able to not change.
Fig. 6 explanation is according to coding/decoding method of the present invention, in being called an image of present image under the situation of obliterated data, the image sheet that allows the described present image of decoding not influenced by loss of data even the image head of described present image is marked between transmission period is lost.In this coding/decoding method, receive the vision signal of H.263 encoding according to standard, wherein the GFID field is filled according to aforementioned coding method.
According to routine, index n and i allow the time situation of coded image in the following description.Mark below using:
Ima (n): the coded image (that is present image) of row n,
Ima (n-i): the coded image of row (n-i), i is integer and i＞0,
GFID (n): be the value of the GFID field of image I ma (n),
GFID (n-i): be the value of the GFID field of image I ma (n-i),
The image leader of image_header (n): Ima (n),
The sheet leader of the image of slice_header (n): Ima (n).
This method comprises and detecting and initialized step 601, its detect with identifier field GFID with the corresponding to mode of filling and be the parameter of initialization coding/decoding method so that explanation GFID field 1 with 2 consistent with coding method whether.This step 601 sends the coding/decoding method of the Proprietary Information of coding method to the used filling mode of reflection preferably in the execution that begins that sends vision signal by coding method thereupon.Predicting under the situation of single mode, do not carry out this initialization step, and use default value corresponding to this mode.In another variant of the present invention,, send a new proprietary signal that reflects the value of described new mode to coding/decoding method, so that the latter can correct interpretation GFID field for allowing to shift gears in the communication process between coding method and coding/decoding method.
Regularly, investigate encoded video signal investigating step 602, so that be identified at the existence of the GFID field in the semanteme of vision signal of coding.When sign during a GFID field, the storage in storing step 603 of its value, and the value of the image head target field of the image that belongs to of described GFID field.Always depend on whether the GFID field is identical to same image, the single value of GFID of storing each image is enough, and no matter the number of the image sheet of each image encoded.The GFID field of described storage and image leader extract from an image encoded of row (n-i), described image preferably can't help fully data lose influence and its time sequencing is positioned at described present image front.So it belongs to identifier field GFID (n-i) and the image leader image_header (n-i) of stored image I ma (n-i).
Under the situation of part or all of obliterated data, influence is greater than the coded image of the row of (n-i), and the image leader of image I ma (n), described coding/decoding method comprises step 604, be used for being called synchronous decoding method on the image sheet leader that weighs synchronous sheet at one again this heavy generation in the described present image Ima (n) that will decode synchronously.Synchronously heavy for carrying out, analyze the data of forming coded image Ima (n), up to a sign code (GBSC or SSC) in the coding semantics, the beginning of an image sheet of indication is not influenced by loss of data.
Described heavy synchronous sheet comprises sheet leader slice_header (n), it self particularly including field GFID (n).
In step 605, compare GFID (n) and value (n-i).
Under the situation of GFID (n)=GFID (n-i), described coding/decoding method is divided into two branches according to the mode of filling the GFID field.Each branch of this method allows the decision decoding parametric, they are present among the image leader image_header (n) usually, if this image leader is not lost, it is essential that these parameters are not subjected to the image sheet of the Ima (n) that loss of data influences for decoding, or for the decoding performance of optimizing the vision signal that is encoded be essential:
1) mode 1: consider two kinds of situations according to Ima (n-i) and Ima (n) type, image type is provided by the value of the position 1 of GFID (n-i) and GFID (n) field.
-a) Ima (n) and Ima (n-i) are the nonanticipating type: step 606 allows field PTYPE (and the PLUSPTYPE by refetching at image leader image_header (n-i), the decoding parametric of the image sheet of the field decision Ima (n) that comprises if you are using)
-b) Ima (n) and Ima (n-i) are type of prediction: step 606 allows field PTYPE (and the PLUSPTYPE by refetching at image leader image_header (n-i), the decoding parametric of the image sheet of the field that comprises if you are using) decision Ima (n) is the same hypothesis among value maintenance and the Ima (n) of position 3-4-5 of the value of position 2 of GFID (n) and the PTYPE field that must be Ima (n-i) except the value of the RTYPE field of Ima (n).
2) mode 2: the type according to Ima (n-i) and Ima (n) is considered two kinds of situations, and image type is provided by the value of the position 1 of GFID (n-i) and GFID (n) field.
-a) Ima (n) and Ima (n-i) are the nonanticipating type: step 607 allow by the field PTYPE that refetches at leader image_header (n-i) (and PLUSPTYPE, the decoding parametric of the image sheet of the field decision Ima (n) that comprises in if you are using),
-b) Ima (n) and Ima (n-i) are type of prediction: with 2) and a), that is step 607 allows field PTYPE (and the PLUSPTYPE by refetching at leader image_header (n-i), the decoding parametric of the image sheet of the field decision Ima (n) that comprises if you are using), have the RTYPE field in addition, it must be fixed on an arbitrary value by coding/decoding method.
Under the situation of GFID (n) ≠ GFID (n-i), described coding/decoding method is divided into two branches according to the mode of filling the GFID field.Each branch of this method allows the decision decoding parametric, they are present among the image leader image_header (n) usually, if this leader is not lost, it is essential that these parameters are not subjected to the image sheet of the Ima (n) that loss of data influences for decoding, or for the decoding performance of optimizing the vision signal that is encoded be essential:
3) mode 1: consider three kinds of situations according to Ima (n-i) and Ima (n) type, image type is provided by the value of the position 1 of GFID (n-i) and GFID (n) field.
-a) Ima (n) and Ima (n-i) are type of prediction: step 608 allows field PTYPE (and the PLUSPTYPE by refetching at leader image_header (n-i), the decoding parametric of the image sheet of the field that comprises if you are using) decision Ima (n) is except the value of the RTYPE field of Ima (n) is the value of the position 2 of GFID (n).Yet, the value of position 3-4-5 that must do the PTYPE field of Ima (n-i) keep with Ima (n) in the same hypothesis.
-b) Ima (n) is that type of prediction and Ima (n-i) are the nonanticipating type: with 3) and a), that is step 608 allows field PTYPE (and the PLUSPTYPE by refetching at leader image_header (n-i), the decoding parametric of the image sheet of the field that comprises if you are using) decision Ima (n) is except the value of the RTYPE field of Ima (n) is the value of the position 2 of GFID (n).Yet, the value of position 3-4-5 that must do the PTYPE field of Ima (n-i) keep with Ima (n) in the same hypothesis.
-c) Ima (n) is that non-type of prediction and Ima (n-i) are type of prediction: step 608 allows (and PLUSPTYPE, the decoding parametric of the image sheet of the field decision Ima (n) that comprises in if you are using) by the field PTYPE that refetches at leader image_header (n-i).Yet, the value of position 3-4-5 that must do the PTYPE field of Ima (n-i) keep with Ima (n) in the same hypothesis.Needn't consider the position 2 of GFID (n).
4) mode 2: the type according to Ima (n-i) and Ima (n) is considered four kinds of situations, and image type is provided by the value of the position 1 of GFID (n-i) and GFID (n) field.
-a) Ima (n) and Ima (n-i) are type of prediction: step 609 does not allow (and PLUSPTYPE, the decoding parametric of the image sheet of the field decision Ima (n) that comprises in if you are using) by the field PTYPE that refetches at leader image_header (n-i).Really, not only be illustrated in this case between Ima (n) and the Ima (n-i) and lose a non-prediction pictures, and presentation format changes the decoding of obstruction Ima (n).Can not repair loss corresponding to data in this case.On the contrary, advantageously use detection in this case to trigger the request of coding/decoding method rapidly, so that ask coding method to send a nonanticipating refreshed image.
-b) Ima (n) and Ima (n-i) are the nonanticipating type: presentation format changes in this case.Step 609 allows field PTYPE (and the PLUSPTYPE by refetching at leader image_header (n-i), the decoding parametric of the image sheet of the field that comprises if you are using) decision Ima (n), the form that uses except the image sheet for decoding Ima (n) are that the value from the position 2 of GFID (n) derives.
-c) Ima (n) is that type of prediction and Ima (n-i) are the nonanticipating type: if the position 2 of GFID (n) and GFID (n-i) is different, then situation 4) and a) generation once more.On the contrary, if the position 2 of GFID (n) and GFID (n-i) is identical, then situation 3) a) take place once more.
-d) Ima (n) is that non-type of prediction and Ima (n-i) are type of prediction: if the position 2 of GFID (n) and GFID (n-i) is different, and this presentation format change.Step 609 allows field PTYPE (and the PLUSPTYPE by refetching at leader image_header (n-i), the decoding parametric of the image sheet of the field that comprises if you are using) decision Ima (n), the form that uses except the image sheet for decoding Ima (n) are that the value from the position 2 of GFID (n) derives.If the position of GFID (n) and GFID (n-i) is 2 identical, step 609 allows (and PLUSPTYPE, the decoding parametric of the image sheet of the field decision Ima (n) that comprises in if you are using) by the field PTYPE that refetches at leader image_header (n-i).
Therefore, can (and PLUSPTYPE be extracted as the described heavy synchronous sheet of decoding in if you are using) and belongs to the essential field of subsequent image sheet of Ima (n) from the field PTYPE of the leader image_header (n-i) of storage.It is possible that the decoding that should be noted that described subsequent image sheet self is not subjected on the loss of data effect at them, and under this kind situation, coding/decoding method will restart from step 602.In other words, except that relevant position 1 by GFID (n) with the information of present image type, the form of present image or by the position 2 of GFID (n) with the type of rounding, other coding characteristic for the required present image Ima (n) of its decoding, the knowledge of the standard the encoding option H.263 that uses during this vision signal of coding that particularly must consider is extracted from the image leader of image I ma (n-i).
Consider that seemingly the coding characteristic of the image of early decoding can be used to the present image of decoding according under aforesaid way 1 and the 2 possible different situations, if the comparison of identifier 605 does not disclose:
The image of-described early decoding and described present image be type of prediction and have different-format,
-or the image of early decoding be non-type of prediction, be type of prediction and have different forms with described present image.
Really, the present image of the type of prediction of only may decoding is only worked as by an image support with early decoding of same form.Memory can have described second form from first form by transmission to the transition of second form and realize especially as a non-prediction pictures of its form.
On the contrary, if the comparison step of identifier 605 discloses:
The image of-described early decoding and described present image be type of prediction and have different-format,
-or the image type of described early decoding be non-type of prediction, be type of prediction and have different forms with described present image,
Before coding/decoding method attempted to decode present image, described coding/decoding method can advantageously be asked the image of a nonanticipating type in this case from coding method.
At last, after according to one of step 606-607-608-609 decision decoding parametric, decoding step 610 uses described decoding parametric according to the sheet of the coding/decoding method decoding present image Ima (n) of standard H.263.
Fig. 7 schematically illustrates of the present invention first and uses, and it is a communication system, comprises coded system 701 and decode system 702, and they are by transmission channel 703 swap datas, and transmission channel 703 for example is wireless or telegraphic style.
About carrying out the equipment of this Code And Decode method, use signal processor especially as much as possible in the coded system level with on the decode system level, these processors are carried out aforementioned different step by the many groups of instructions that execution is stored in the memory to vision signal.
Fig. 8 represents that of the present invention second uses, and realizes a portable telephone 801 in wherein that described coding method and coding/decoding method, described portable telephone provide to transmit with telephone set of the same type and the possibility of receiving function.The integrated video camera 802 of this telephone set provides original uncoded vision signal, for example allows the people who has it to be taken.This telephone set comprises that one acts on encoding device according to the present invention simultaneously according to the encoder of standard H.263, so that the vision signal of coding is provided, the reflection of the GFID field of described vision signal is according to the feature of coding method of the present invention.
When this telephone set is operated in accepting state, after receiving by vision signal according to the coding that is operated in the telephone set generation of transmission state and transmits according to the present invention, obliterated data causes losing under the situation of image sheet of present image during the transmission, H.263, the image sheet that described present image is not influenced by loss of data is decoded according to standard by decoder, wherein by using the coding characteristic lost and to have recovered by coding/decoding method according to the present invention.Decoded image is presented on the screen 803.
The telephone set of this type is advantageously used in the environment of transfer of data, particularly according to the data of the video type of standard code H.263, by the transmission network of UMTS type.
Claims (8)
1. the image sequence that is used to encode provides the method for the sequence of an image encoded, each image encoded comprises an image leader and is cut into and respectively comprises a plurality of image sheets of head target, each sheet leader comprises the identifier field that will be identified, this identifier field is identical for all sheets of a given coded image, the image head target sign of the image of current and previous coding relates to the sign of the identifier field of these two kinds of images, the image leader of the image of present encoding changes the change of the identifier field of the image that relates to described present encoding with respect to the image head target of the image of previous coding, it is characterized in that, thereby described method comprises that filling identifier field with value makes the step of a coding characteristic of image of each reflection present encoding of this identifier field.
2. method that the image of a present encoding of the part of the sequence that forms an image encoded is decoded, each image encoded is comprised an image leader by expectation, and be cut into and respectively comprise a plurality of of head targets, each sheet leader comprises the identifier field that will be identified, a coding characteristic of the image of each reflection present encoding of described identifier field, it is characterized in that described coding/decoding method may further comprise the steps:
The identifier field of the identifier field of-more previous decoded image and the image of present encoding,
The coding characteristic that-the identifier field of image by using present encoding is discerned, and the coding characteristic of being discerned by the image leader that uses described previous decoded image comes the image of described present encoding is decoded, and its condition is relatively not disclosing of described identifier field:
I) image of described previous decoded image and described present encoding is that type of prediction and they have different forms,
J) or described previous decoded image be that nonanticipating type and the image of described present encoding are that type of prediction and they have different forms.
3. method as claimed in claim 1 or 2, wherein, second of the image type of the image that first reflection of identifier field is encoded and identifier field is reflected as the rounding type that coded image uses.
4. method as claimed in claim 1 or 2, wherein, the picture format of the image that the image type of the image that first reflection of identifier field is encoded and second reflection of identifier field are encoded.
5. the image sequence that is used to encode provides the system of the sequence of an image encoded, each image encoded comprises an image leader and is cut into and respectively comprises a plurality of image sheets of head target, each sheet leader comprises the identifier field that will be identified, this identifier field is identical for all sheets of a given coded image, the image head target sign of the image of current and previous coding relates to the sign of the identifier field of these two kinds of images, the image leader of the current image that is encoded changes the change of the identifier field that relates to the described current image that is encoded with respect to the image head target of the image of previous coding, it is characterized in that, thereby the described system that is used to encode comprises that filling identifier field with value makes the device of a coding characteristic of image of each reflection present encoding of this identifier field.
6. system that current encoded image is decoded to the part of the sequence that forms an image encoded, each comprises an identifier field described image encoded, a coding characteristic of the current image that is encoded of each reflection of described identifier field, it is characterized in that described decode system comprises:
-be used for the device of the identifier field of the identifier field of more previous decoded image and the current image that is encoded,
-be used for device that the image of described present encoding is decoded, it is used for the coding characteristic discerned by the identifier field of the image that uses present encoding, and the coding characteristic of being discerned by the image leader that uses described previous decoded image comes the image of described present encoding is decoded, and its condition is relatively not disclosing of described identifier field:
I) image of described previous decoded image and described present encoding is that type of prediction and they have different forms,
J) or described previous decoded image be that nonanticipating type and the image of described present encoding are that type of prediction and they have different forms.
7. one kind comprises as coded system as described in the claim 5 with as the communication system of decode system as described in the claim 6.
8. one kind comprises as coded system as described in the claim 5 with as the portable telephone of decode system as described in the claim 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB02143736XA CN100375520C (en) | 2002-08-22 | 2002-08-22 | Video coding and decoding method for reducing data loss in transmission of coded video signals iva channel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB02143736XA CN100375520C (en) | 2002-08-22 | 2002-08-22 | Video coding and decoding method for reducing data loss in transmission of coded video signals iva channel |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1477875A CN1477875A (en) | 2004-02-25 |
| CN100375520C true CN100375520C (en) | 2008-03-12 |
Family
ID=34148324
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB02143736XA Expired - Lifetime CN100375520C (en) | 2002-08-22 | 2002-08-22 | Video coding and decoding method for reducing data loss in transmission of coded video signals iva channel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100375520C (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101213843B (en) * | 2005-06-30 | 2010-09-29 | 法国电信公司 | Video coding method and device |
| CN101540901B (en) * | 2008-03-20 | 2011-04-06 | 华为技术有限公司 | Method and device for coding/decoding |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5847767A (en) * | 1994-03-25 | 1998-12-08 | Nec Corporation | Compact image transmission system and image reproducing device capable of reproducing a DCT-based code at a high speed without deterioration of picture quality |
| CN1293519A (en) * | 1999-10-19 | 2001-05-02 | 三星电子株式会社 | Digital video frequency encoding method and equipment |
| WO2001050766A1 (en) * | 1999-12-22 | 2001-07-12 | Neomtel Co., Ltd. | A system for moving image data using wireless communication and the method of the same |
-
2002
- 2002-08-22 CN CNB02143736XA patent/CN100375520C/en not_active Expired - Lifetime
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5847767A (en) * | 1994-03-25 | 1998-12-08 | Nec Corporation | Compact image transmission system and image reproducing device capable of reproducing a DCT-based code at a high speed without deterioration of picture quality |
| CN1293519A (en) * | 1999-10-19 | 2001-05-02 | 三星电子株式会社 | Digital video frequency encoding method and equipment |
| WO2001050766A1 (en) * | 1999-12-22 | 2001-07-12 | Neomtel Co., Ltd. | A system for moving image data using wireless communication and the method of the same |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1477875A (en) | 2004-02-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110677672B (en) | Method and system for encoding video with overlay | |
| US8605779B2 (en) | Mechanisms to conceal real time video artifacts caused by frame loss | |
| EP3355584A1 (en) | Video encoding method, video decoding method, video encoding apparatus, video decoding apparatus, video encoding program and video decoding program | |
| US6137912A (en) | Method of multichannel data compression | |
| NO321859B1 (en) | Method for decoding signal for live image and decoding device for signal for live image using the method | |
| US6961082B2 (en) | Image display control system reducing image transmission delay | |
| CN111131817A (en) | Screen sharing method, device, storage medium and screen sharing system | |
| EP3951766B1 (en) | Image display control device, transmission device, image display control method, and program | |
| US20120106632A1 (en) | Method and apparatus for error resilient long term referencing block refresh | |
| CN115134629A (en) | Video transmission method, system, device and storage medium | |
| CN100375520C (en) | Video coding and decoding method for reducing data loss in transmission of coded video signals iva channel | |
| JP4855622B2 (en) | Video encoding / decoding method for reducing data loss in an encoded video signal transmitted over a channel | |
| JP3493985B2 (en) | Video communication evaluation device | |
| JP2002010265A (en) | Transmitting apparatus and method and receiving apparatus and method | |
| CN110636293A (en) | Video encoding method, video decoding method, video encoding apparatus, video decoding apparatus, storage medium, and electronic apparatus | |
| US8731045B2 (en) | Motion encoding and decoding | |
| KR100574867B1 (en) | Mobile communication terminal | |
| CN110971870B (en) | Data processing method for image display | |
| CN101494718A (en) | Method and apparatus for encoding image | |
| JP2003125410A (en) | Image encoding device, image decoding device and method, image encoding program, image decoding program | |
| JP3226637B2 (en) | Coding device and decoding device for predictive coding method | |
| US10567802B2 (en) | Method for optimising the transmission of video data streams in a wireless network | |
| CN113923453B (en) | Time domain filtering method and device for video, electronic equipment and storage medium | |
| JP2003125409A (en) | Image encoding device, image decoding device and method, image encoding program, image decoding program | |
| CN110519599B (en) | Video coding method and device based on distributed analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CX01 | Expiry of patent term |
Granted publication date: 20080312 |
|
| CX01 | Expiry of patent term |